ipynb/robin_li-firm_level_innovation_and_ceo_compensation.ipynb

{
    "cells": [
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "# Firm Level Innovation and CEO Compensation\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Using the panel data on US patents from 1992-2006 and the data on granted stock options value for the Chief Executive Office (CEO) in matched US firms, this study aims to investigate the relationship between the value of CEO’s granted stock options and the firm-level innovation activity, measured by number of patent granted each year.This study aims to provide empirical evidence for Manso(2011)’s theory, which predicts that\n",
                "the optimal-motivating incentive scheme exhibits a substantial tolerance for the early failure,\n",
                "the reward for long-term success and the commitment to a long-term compensation plan for\n",
                "CEOs by board of directors"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 1,
            "metadata": {},
            "outputs": [],
            "source": [
                "import matplotlib.pyplot as plt\n",
                "import numpy as np\n",
                "import pandas as pd\n",
                "import qeds\n",
                "\n",
                "from sklearn.linear_model import LinearRegression\n",
                "import seaborn as sns\n",
                "\n",
                "%matplotlib inline"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Data Cleaning"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "The primary data in this study comes from the [National Bureau of Economic Research (NBER) Patent Database](https://sites.google.com/site/patentdataproject/Home/downloads) and Wharton Research Data Service (WRDS) Compustat Executive Compensation Database."
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Patent Data \n",
                "\n",
                "Patent data `patsic06_mar09_ip.csv` contains informtion regarding the application and granted year, US patent class, along with all the original IPCs for the patent, the original assignee number, and all the pdpass numbers that result from splitting jointly owned patents, standardizing names, etc."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 2,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>appyear</th>\n",
                            "      <th>assignee</th>\n",
                            "      <th>cat</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>icl_class</th>\n",
                            "      <th>icl_maingroup</th>\n",
                            "      <th>nclass</th>\n",
                            "      <th>patent</th>\n",
                            "      <th>pdpass</th>\n",
                            "      <th>subcat</th>\n",
                            "      <th>subclass</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1974</td>\n",
                            "      <td>251415.0</td>\n",
                            "      <td>6.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>A41D</td>\n",
                            "      <td>19.0</td>\n",
                            "      <td>2</td>\n",
                            "      <td>3930271</td>\n",
                            "      <td>10030271.0</td>\n",
                            "      <td>63.0</td>\n",
                            "      <td>161.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1974</td>\n",
                            "      <td>246000.0</td>\n",
                            "      <td>6.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>A47D</td>\n",
                            "      <td>7.0</td>\n",
                            "      <td>5</td>\n",
                            "      <td>3930272</td>\n",
                            "      <td>10156902.0</td>\n",
                            "      <td>65.0</td>\n",
                            "      <td>11.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1975</td>\n",
                            "      <td>10490.0</td>\n",
                            "      <td>6.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>A47C</td>\n",
                            "      <td>21.0</td>\n",
                            "      <td>5</td>\n",
                            "      <td>3930273</td>\n",
                            "      <td>10112031.0</td>\n",
                            "      <td>65.0</td>\n",
                            "      <td>331.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1975</td>\n",
                            "      <td>10490.0</td>\n",
                            "      <td>6.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>A47D</td>\n",
                            "      <td>7.0</td>\n",
                            "      <td>5</td>\n",
                            "      <td>3930273</td>\n",
                            "      <td>10112031.0</td>\n",
                            "      <td>65.0</td>\n",
                            "      <td>331.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1974</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>B63B</td>\n",
                            "      <td>7.0</td>\n",
                            "      <td>114</td>\n",
                            "      <td>3930274</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>55.0</td>\n",
                            "      <td>2.0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   appyear  assignee  cat  gyear icl_class  icl_maingroup  nclass   patent  \\\n",
                            "0     1974  251415.0  6.0   1976      A41D           19.0       2  3930271   \n",
                            "1     1974  246000.0  6.0   1976      A47D            7.0       5  3930272   \n",
                            "2     1975   10490.0  6.0   1976      A47C           21.0       5  3930273   \n",
                            "3     1975   10490.0  6.0   1976      A47D            7.0       5  3930273   \n",
                            "4     1974       0.0  5.0   1976      B63B            7.0     114  3930274   \n",
                            "\n",
                            "       pdpass  subcat  subclass  \n",
                            "0  10030271.0    63.0     161.0  \n",
                            "1  10156902.0    65.0      11.0  \n",
                            "2  10112031.0    65.0     331.0  \n",
                            "3  10112031.0    65.0     331.0  \n",
                            "4         NaN    55.0       2.0  "
                        ]
                    },
                    "execution_count": 2,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "import csv\n",
                "pats = pd.read_csv(\"/home/jupyter/494 Thesis/patsic06_mar09_ip.csv\")\n",
                "pats.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Variable Description\n",
                "  \n",
                "* Patent data\n",
                "  \n",
                "`appyear` records the year when the patent is filed and applied; `gyear` is the year when a patent is granted. Each granted patent has a unique patent number as `patent`.  \n",
                "  \n",
                "`cat`, `icl_class`, `icl_maingroup`, `nclass`, `subcat`, `subclass` are patent class or technology class of each patent, which are less relevant in my study\n",
                "  \n",
                "* Assignee identifier\n",
                "  \n",
                "`assignee` is the number assigned to a company when applying or granting a patent, originally created by the United States Patent Office (U.S.P.O). `pdpass` is also a unique assignee number and it is used to match data to the Compustat. One `pdpass` might be assigned by multiple `assignee` numbers \n",
                "\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "  \n",
                "#### Treatment of Unmatched assignee \n",
                "  \n",
                "There are patents which are not matched to the assignee by the name matching algorith developed by the NBER Patent database, these observations are shown as `assignee = 0` or `pdpass = NaN`. These observation are automatically dropped "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 3,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>pdpass</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>6361394.0</td>\n",
                            "      <td>1979</td>\n",
                            "      <td>6</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>6363799.0</td>\n",
                            "      <td>1976</td>\n",
                            "      <td>4</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>6363799.0</td>\n",
                            "      <td>1977</td>\n",
                            "      <td>3</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>6363799.0</td>\n",
                            "      <td>1978</td>\n",
                            "      <td>4</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>6363799.0</td>\n",
                            "      <td>1979</td>\n",
                            "      <td>5</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "      pdpass  gyear  patent_count\n",
                            "0  6361394.0   1979             6\n",
                            "1  6363799.0   1976             4\n",
                            "2  6363799.0   1977             3\n",
                            "3  6363799.0   1978             4\n",
                            "4  6363799.0   1979             5"
                        ]
                    },
                    "execution_count": 3,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# observations with pdpass not missing \n",
                "pats = pats[pats['pdpass'].notnull()]\n",
                "\n",
                "# count number of unique patent numbers to have patent count for a assignee in a year\n",
                "pat_yc1 = pats.groupby([\"pdpass\",\"gyear\"])[\"patent\"].nunique().to_frame().reset_index()\\\n",
                ".rename(columns={\"patent\":\"patent_count\"})\n",
                "\n",
                "pat_yc1.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 4,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "<class 'pandas.core.frame.DataFrame'>\n",
                        "RangeIndex: 570089 entries, 0 to 570088\n",
                        "Data columns (total 3 columns):\n",
                        "pdpass          570089 non-null float64\n",
                        "gyear           570089 non-null int64\n",
                        "patent_count    570089 non-null int64\n",
                        "dtypes: float64(1), int64(2)\n",
                        "memory usage: 13.0 MB\n"
                    ]
                }
            ],
            "source": [
                "pat_yc1.info()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Add `gvkey`\n",
                "  \n",
                "`gvkey` is Wharton Rearch Data Service (WRDS)'s identifier of companies\n",
                "\n",
                "`dynasscsv.csv` keeps record of all dynamic change in patent ownership. Change in patent ownership may arise from merge and acquisition of companies, or sale of patent. In this study, I only focus on the original assignee (as `pdpco1` and `gvkey1`) of the patent and disregard any dynamic change in patent ownership. This is because I am more interested in the relationship between patent and compensation through executive's decision on company's own research project and the process of exploring new ideas, not through purchasing patent or acquire patents from other companies, as patent purchase is a decision with known payoffs      "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 5,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>pdpass</th>\n",
                            "      <th>pdpco</th>\n",
                            "      <th>gvkey</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>-19702.0</td>\n",
                            "      <td>66448</td>\n",
                            "      <td>66448</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>-19541.0</td>\n",
                            "      <td>121844</td>\n",
                            "      <td>121844</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>-19540.0</td>\n",
                            "      <td>121844</td>\n",
                            "      <td>121844</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>-19291.0</td>\n",
                            "      <td>62613</td>\n",
                            "      <td>62613</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>-19289.0</td>\n",
                            "      <td>176118</td>\n",
                            "      <td>176118</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "    pdpass   pdpco   gvkey\n",
                            "0 -19702.0   66448   66448\n",
                            "1 -19541.0  121844  121844\n",
                            "2 -19540.0  121844  121844\n",
                            "3 -19291.0   62613   62613\n",
                            "4 -19289.0  176118  176118"
                        ]
                    },
                    "execution_count": 5,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "dynass = pd.read_csv(\"/home/jupyter/494 Thesis/dynasscsv.csv\")\n",
                "dynass = dynass[[\"pdpass\",\"pdpco1\",\"gvkey1\"]] \n",
                "dynass = dynass.rename(columns = {\"pdpco1\":\"pdpco\", \"gvkey1\":\"gvkey\"})\n",
                "dynass = dynass.astype({\"pdpass\": float})\n",
                "dynass.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "In many cases, `pdpass` equals `gvkey` for a company. My goal is to have `gvkey` as the unqie identifier for companies, thus I drop any `pdpass` that is not matched to a `gvkey`, thus I perform the inner merge. In addition, we would like to investigate any change in patent count after executives granted any stock options; since data range for executives' stock option is from 1992-2006, we would only focus on the subset of patent count that is after 1992  "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 6,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>pdpass</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>pdpco</th>\n",
                            "      <th>gvkey</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>10030467.0</td>\n",
                            "      <td>2006</td>\n",
                            "      <td>1</td>\n",
                            "      <td>8006</td>\n",
                            "      <td>8006</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>18</th>\n",
                            "      <td>10030734.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>4</td>\n",
                            "      <td>1847</td>\n",
                            "      <td>1847</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>19</th>\n",
                            "      <td>10030734.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>5</td>\n",
                            "      <td>1847</td>\n",
                            "      <td>1847</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>20</th>\n",
                            "      <td>10030734.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>2</td>\n",
                            "      <td>1847</td>\n",
                            "      <td>1847</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>21</th>\n",
                            "      <td>10030734.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>4</td>\n",
                            "      <td>1847</td>\n",
                            "      <td>1847</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "        pdpass  gyear  patent_count  pdpco  gvkey\n",
                            "4   10030467.0   2006             1   8006   8006\n",
                            "18  10030734.0   1992             4   1847   1847\n",
                            "19  10030734.0   1994             5   1847   1847\n",
                            "20  10030734.0   1995             2   1847   1847\n",
                            "21  10030734.0   1996             4   1847   1847"
                        ]
                    },
                    "execution_count": 6,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pgvkey = pd.merge(pat_yc1, dynass, on=\"pdpass\",how=\"inner\")\n",
                "\n",
                "# align with executive compensation data (1992 onwards)\n",
                "pgvkey1 = pgvkey.loc[pgvkey[\"gyear\"]>=1992]\n",
                "pgvkey1.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Treating Missing Year's Patent Count\n",
                "  \n",
                "For each company (as identified by gvkey), the reason for missing value in the patent count can be: 1) there might be patents granted in that year, however they are not matched to the pdpass, and thus not to the company in the NBER patent database; 2) there is simply no patent granted in that year. As a result, the rule of thumb of treating patent count missing value each year is simply replace it with 0 in this study  "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 7,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th>gyear</th>\n",
                            "      <th>1992</th>\n",
                            "      <th>1993</th>\n",
                            "      <th>1994</th>\n",
                            "      <th>1995</th>\n",
                            "      <th>1996</th>\n",
                            "      <th>1997</th>\n",
                            "      <th>1998</th>\n",
                            "      <th>1999</th>\n",
                            "      <th>2000</th>\n",
                            "      <th>2001</th>\n",
                            "      <th>2002</th>\n",
                            "      <th>2003</th>\n",
                            "      <th>2004</th>\n",
                            "      <th>2005</th>\n",
                            "      <th>2006</th>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>gvkey</th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "      <th></th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>1004</th>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1010</th>\n",
                            "      <td>3.0</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>6.0</td>\n",
                            "      <td>2.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1012</th>\n",
                            "      <td>NaN</td>\n",
                            "      <td>2.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>2.0</td>\n",
                            "      <td>2.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>2.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1013</th>\n",
                            "      <td>9.0</td>\n",
                            "      <td>15.0</td>\n",
                            "      <td>12.0</td>\n",
                            "      <td>12.0</td>\n",
                            "      <td>9.0</td>\n",
                            "      <td>10.0</td>\n",
                            "      <td>10.0</td>\n",
                            "      <td>23.0</td>\n",
                            "      <td>23.0</td>\n",
                            "      <td>47.0</td>\n",
                            "      <td>69.0</td>\n",
                            "      <td>100.0</td>\n",
                            "      <td>97.0</td>\n",
                            "      <td>82.0</td>\n",
                            "      <td>83.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1016</th>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "gyear  1992  1993  1994  1995  1996  1997  1998  1999  2000  2001  2002  \\\n",
                            "gvkey                                                                     \n",
                            "1004    1.0   3.0   1.0   3.0   NaN   NaN   NaN   NaN   1.0   NaN   NaN   \n",
                            "1010    3.0   3.0   6.0   2.0   1.0   NaN   NaN   NaN   NaN   NaN   NaN   \n",
                            "1012    NaN   2.0   NaN   2.0   2.0   NaN   1.0   NaN   NaN   3.0   NaN   \n",
                            "1013    9.0  15.0  12.0  12.0   9.0  10.0  10.0  23.0  23.0  47.0  69.0   \n",
                            "1016    NaN   NaN   NaN   NaN   1.0   NaN   NaN   NaN   1.0   NaN   NaN   \n",
                            "\n",
                            "gyear   2003  2004  2005  2006  \n",
                            "gvkey                           \n",
                            "1004     1.0   NaN   1.0   NaN  \n",
                            "1010     NaN   NaN   NaN   NaN  \n",
                            "1012     3.0   NaN   NaN   2.0  \n",
                            "1013   100.0  97.0  82.0  83.0  \n",
                            "1016     NaN   NaN   NaN   NaN  "
                        ]
                    },
                    "execution_count": 7,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pgvkey2 = pgvkey1[[\"gvkey\",\"gyear\",\"patent_count\"]]\n",
                "pgvkey3 = pgvkey2.pivot_table(index=\"gvkey\", columns =\"gyear\", values=\"patent_count\",\n",
                "                             aggfunc= \"sum\")\n",
                "pgvkey3.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 8,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count\n",
                            "0   1004   1992           1.0\n",
                            "1   1004   1993           3.0\n",
                            "2   1004   1994           1.0\n",
                            "3   1004   1995           3.0\n",
                            "4   1004   1996           NaN"
                        ]
                    },
                    "execution_count": 8,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pgvkey3 = pgvkey3.stack(\"gyear\",dropna=False).to_frame().reset_index()\\\n",
                "    .rename(columns={0: 'patent_count'})\n",
                "pgvkey3.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 9,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count\n",
                            "0   1004   1992           1.0\n",
                            "1   1004   1993           3.0\n",
                            "2   1004   1994           1.0\n",
                            "3   1004   1995           3.0\n",
                            "4   1004   1996           0.0"
                        ]
                    },
                    "execution_count": 9,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pgvkey3[\"patent_count\"] = pgvkey3[\"patent_count\"].fillna(0)\n",
                "pgvkey3.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 10,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "<class 'pandas.core.frame.DataFrame'>\n",
                        "RangeIndex: 88365 entries, 0 to 88364\n",
                        "Data columns (total 3 columns):\n",
                        "gvkey           88365 non-null int64\n",
                        "gyear           88365 non-null int64\n",
                        "patent_count    88365 non-null float64\n",
                        "dtypes: float64(1), int64(2)\n",
                        "memory usage: 2.0 MB\n"
                    ]
                }
            ],
            "source": [
                "pgvkey3.info()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Compustate Executive Compensation - Stock Option Granted\n",
                "\n",
                "`stock_option.csv` contains data on the Black-Sholes value of stock option granted by the company's executives 1992-2006. It has two sets of variables, one describes the executive information, such as:\n",
                "\n",
                "* reload = reload option (1 = if executive exercised reload option)\n",
                "* year = year when stock option is granted \n",
                "* pceo = description of the executive's position\n",
                "* blkshval = Black-Scholes value of the stock option ...\n",
                "\n",
                "Another set of variables describes the charateristics of the company, for example:\n",
                "* value = company value\n",
                "* ticker = company ticker\n",
                "* sic = Standard Industry Classification code ...\n",
                "  \n",
                "As I will show later, in this study, I willl only look into the stock options value granted by the CEOs of the company (as categorized in `pceo` variable). I will sum the *Black-Scholes values* of the stock option in a given year for a CEO (plus reloaded stock option values, if any) to produce the measure of CEO's incentive compensation"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 11,
            "metadata": {},
            "outputs": [],
            "source": [
                "stockog= pd.read_csv(\"/home/jupyter/494 Thesis/stock_option.csv\")\n",
                "\n",
                "# change column names to lower case\n",
                "stockog.columns = [x.lower() for x in stockog.columns]"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 12,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>mktpric</th>\n",
                            "      <th>expric</th>\n",
                            "      <th>pcdvalue</th>\n",
                            "      <th>value</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>year</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>60.000</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>13.750</td>\n",
                            "      <td>13.750</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>518.838</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>20050117.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>75.000</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>17.500</td>\n",
                            "      <td>17.500</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>825.424</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>20051011.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>17</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>200.000</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>20.000</td>\n",
                            "      <td>20.000</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2515.578</td>\n",
                            "      <td>1133.373</td>\n",
                            "      <td>20060709.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>18</th>\n",
                            "      <td>1</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>2</td>\n",
                            "      <td>7.768</td>\n",
                            "      <td>1.24</td>\n",
                            "      <td>25.875</td>\n",
                            "      <td>25.875</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>43.316</td>\n",
                            "      <td>37.718</td>\n",
                            "      <td>20001010.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>19</th>\n",
                            "      <td>1</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>3</td>\n",
                            "      <td>5.371</td>\n",
                            "      <td>0.86</td>\n",
                            "      <td>25.875</td>\n",
                            "      <td>25.875</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>38.396</td>\n",
                            "      <td>29.110</td>\n",
                            "      <td>20011009.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "    reload  co_per_rol  grntnum  numsecur  pcttotopt  mktpric  expric  \\\n",
                            "1        0        5623        1    60.000      24.00   13.750  13.750   \n",
                            "6        0        5623        1    75.000      26.05   17.500  17.500   \n",
                            "17       0        5623        1   200.000      32.08   20.000  20.000   \n",
                            "18       1        5623        2     7.768       1.24   25.875  25.875   \n",
                            "19       1        5623        3     5.371       0.86   25.875  25.875   \n",
                            "\n",
                            "     pcdvalue     value  blkshval      exdate  year pceo  gvkey  cusip  \\\n",
                            "1   5 PERCENT   518.838   288.811  20050117.0  1994  CEO   1004  36110   \n",
                            "6   5 PERCENT   825.424   325.076  20051011.0  1995  CEO   1004  36110   \n",
                            "17  5 PERCENT  2515.578  1133.373  20060709.0  1996  CEO   1004  36110   \n",
                            "18  5 PERCENT    43.316    37.718  20001010.0  1996  CEO   1004  36110   \n",
                            "19  5 PERCENT    38.396    29.110  20011009.0  1996  CEO   1004  36110   \n",
                            "\n",
                            "                         sicdesc  \\\n",
                            "1   MACHINERY AND EQUIPMENT-WHSL   \n",
                            "6   MACHINERY AND EQUIPMENT-WHSL   \n",
                            "17  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "18  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "19  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                            naicsdesc ticker     naics   sic  \n",
                            "1   Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "6   Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "17  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "18  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "19  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  "
                        ]
                    },
                    "execution_count": 12,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# having compensation for the CEOs only\n",
                "stockog = stockog.loc[pd.notnull(stockog[\"pceo\"])]\n",
                "stockog.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 13,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  year  blkshval\n",
                            "0   1004  1994   288.811\n",
                            "1   1004  1995   325.076\n",
                            "2   1004  1996  1443.320\n",
                            "3   1004  1997  2528.623\n",
                            "4   1004  1998  1644.472"
                        ]
                    },
                    "execution_count": 13,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# sum of stock option and reload stock option \n",
                "stockog1 = stockog.groupby([\"gvkey\",\"year\"])[\"blkshval\"].sum().to_frame().reset_index()\n",
                "stockog1.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Treating Missing Values"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 14,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  year  blkshval\n",
                            "0   1004  1992       NaN\n",
                            "1   1004  1993       NaN\n",
                            "2   1004  1994   288.811\n",
                            "3   1004  1995   325.076\n",
                            "4   1004  1996  1443.320"
                        ]
                    },
                    "execution_count": 14,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# filling in missing years and keep them as missing\n",
                "\n",
                "stockog2 = stockog1.pivot_table(index=\"gvkey\", columns =\"year\", values=\"blkshval\")\n",
                "stockog2 = stockog2.stack(\"year\",dropna=False).to_frame().reset_index().rename(\n",
                "    columns={0: 'blkshval'})\n",
                "stockog2.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Merged Dataset"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 15,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval\n",
                            "0   1004   1992           1.0  1992       NaN\n",
                            "1   1004   1993           3.0  1993       NaN\n",
                            "2   1004   1994           1.0  1994   288.811\n",
                            "3   1004   1995           3.0  1995   325.076\n",
                            "4   1004   1996           0.0  1996  1443.320"
                        ]
                    },
                    "execution_count": 15,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "patcom = pd.merge(pgvkey3, stockog2, left_on=[\"gvkey\",\"gyear\"], right_on=[\"gvkey\",\"year\"], \n",
                "                  how=\"inner\")\n",
                "patcom.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Restore Firm Charateristics \n",
                "  \n",
                "All company's descriptive variables are restored by merging `stockog1` dataframe with the original dataframe, since these variables will be candiate control variables for dealing with endogeneity problem and omitted variable bias. For example: \n",
                "\n",
                "* value = company value\n",
                "* sic = Standard Industry Classification \n",
                "  \n",
                "  etc...\n",
                "\n",
                "We drop variable `year` and `blkshval` to avoid creating repetitive columns for year and blkshval when merging "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 17,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>mktpric</th>\n",
                            "      <th>expric</th>\n",
                            "      <th>pcdvalue</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>year</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>60.0</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>13.75</td>\n",
                            "      <td>13.75</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>518.838</td>\n",
                            "      <td>20050117.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>17.50</td>\n",
                            "      <td>17.50</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>825.424</td>\n",
                            "      <td>20051011.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>17</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>20.00</td>\n",
                            "      <td>20.00</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2515.578</td>\n",
                            "      <td>20060709.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>53</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>150.0</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>23.50</td>\n",
                            "      <td>23.50</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2216.854</td>\n",
                            "      <td>20070714.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>75</th>\n",
                            "      <td>0</td>\n",
                            "      <td>5623</td>\n",
                            "      <td>1</td>\n",
                            "      <td>195.0</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>23.50</td>\n",
                            "      <td>23.50</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2363.166</td>\n",
                            "      <td>20090101.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>1004</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "    reload  co_per_rol  grntnum  numsecur  pcttotopt  mktpric  expric  \\\n",
                            "1        0        5623        1      60.0      24.00    13.75   13.75   \n",
                            "6        0        5623        1      75.0      26.05    17.50   17.50   \n",
                            "17       0        5623        1     200.0      32.08    20.00   20.00   \n",
                            "53       0        5623        1     150.0      16.84    23.50   23.50   \n",
                            "75       0        5623        1     195.0      23.59    23.50   23.50   \n",
                            "\n",
                            "     pcdvalue     value      exdate  year pceo  gvkey  cusip  \\\n",
                            "1   5 PERCENT   518.838  20050117.0  1994  CEO   1004  36110   \n",
                            "6   5 PERCENT   825.424  20051011.0  1995  CEO   1004  36110   \n",
                            "17  5 PERCENT  2515.578  20060709.0  1996  CEO   1004  36110   \n",
                            "53  5 PERCENT  2216.854  20070714.0  1997  CEO   1004  36110   \n",
                            "75  5 PERCENT  2363.166  20090101.0  1998  CEO   1004  36110   \n",
                            "\n",
                            "                         sicdesc  \\\n",
                            "1   MACHINERY AND EQUIPMENT-WHSL   \n",
                            "6   MACHINERY AND EQUIPMENT-WHSL   \n",
                            "17  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "53  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "75  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                            naicsdesc ticker     naics   sic  \n",
                            "1   Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "6   Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "17  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "53  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  \n",
                            "75  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080  "
                        ]
                    },
                    "execution_count": 17,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "stockog3 = stockog.drop_duplicates(subset=[\"gvkey\",\"year\"], \n",
                "                                   keep=\"first\").drop(columns=[\"blkshval\"])\n",
                "stockog3.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 18,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>pcdvalue</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>60.0</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>518.838</td>\n",
                            "      <td>20050117.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>825.424</td>\n",
                            "      <td>20051011.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2515.578</td>\n",
                            "      <td>20060709.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 22 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "0   1004   1992           1.0  1992       NaN     NaN         NaN      NaN   \n",
                            "1   1004   1993           3.0  1993       NaN     NaN         NaN      NaN   \n",
                            "2   1004   1994           1.0  1994   288.811     0.0      5623.0      1.0   \n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...   pcdvalue     value      exdate  pceo  cusip  \\\n",
                            "0       NaN        NaN  ...        NaN       NaN         NaN   NaN    NaN   \n",
                            "1       NaN        NaN  ...        NaN       NaN         NaN   NaN    NaN   \n",
                            "2      60.0      24.00  ...  5 PERCENT   518.838  20050117.0   CEO  36110   \n",
                            "3      75.0      26.05  ...  5 PERCENT   825.424  20051011.0   CEO  36110   \n",
                            "4     200.0      32.08  ...  5 PERCENT  2515.578  20060709.0   CEO  36110   \n",
                            "\n",
                            "                        sicdesc  \\\n",
                            "0                           NaN   \n",
                            "1                           NaN   \n",
                            "2  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "3  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                           naicsdesc ticker     naics     sic  \n",
                            "0                                                NaN    NaN       NaN     NaN  \n",
                            "1                                                NaN    NaN       NaN     NaN  \n",
                            "2  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "3  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "4  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "\n",
                            "[5 rows x 22 columns]"
                        ]
                    },
                    "execution_count": 18,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat_full = pd.merge(patcom, stockog3, left_on=[\"gvkey\",\"year\"],right_on=[\"gvkey\",\"year\"], how=\"left\")\n",
                "pat_full.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "For some observations, we lost company charatertistics when merging the data, for example `cusip`, `sicdesc` and we would like to restore these details. Here, I would only restore `sic` and use it to classify each company by its industry. I perform by conducting a groupby missing value filling. "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 19,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>pcdvalue</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>60.0</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>518.838</td>\n",
                            "      <td>20050117.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>825.424</td>\n",
                            "      <td>20051011.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>5 PERCENT</td>\n",
                            "      <td>2515.578</td>\n",
                            "      <td>20060709.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 22 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "0   1004   1992           1.0  1992       NaN     NaN         NaN      NaN   \n",
                            "1   1004   1993           3.0  1993       NaN     NaN         NaN      NaN   \n",
                            "2   1004   1994           1.0  1994   288.811     0.0      5623.0      1.0   \n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...   pcdvalue     value      exdate  pceo  cusip  \\\n",
                            "0       NaN        NaN  ...        NaN       NaN         NaN   NaN    NaN   \n",
                            "1       NaN        NaN  ...        NaN       NaN         NaN   NaN    NaN   \n",
                            "2      60.0      24.00  ...  5 PERCENT   518.838  20050117.0   CEO  36110   \n",
                            "3      75.0      26.05  ...  5 PERCENT   825.424  20051011.0   CEO  36110   \n",
                            "4     200.0      32.08  ...  5 PERCENT  2515.578  20060709.0   CEO  36110   \n",
                            "\n",
                            "                        sicdesc  \\\n",
                            "0                           NaN   \n",
                            "1                           NaN   \n",
                            "2  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "3  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                           naicsdesc ticker     naics     sic  \n",
                            "0                                                NaN    NaN       NaN  5080.0  \n",
                            "1                                                NaN    NaN       NaN  5080.0  \n",
                            "2  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "3  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "4  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0  \n",
                            "\n",
                            "[5 rows x 22 columns]"
                        ]
                    },
                    "execution_count": 19,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# groupby filling missing values\n",
                "\n",
                "pat_full['sic'] = pat_full.groupby(['gvkey'])['sic'].ffill()\n",
                "pat_full['sic'] = pat_full.groupby(['gvkey'])['sic'].bfill()\n",
                "\n",
                "pat_full.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 20,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "<class 'pandas.core.frame.DataFrame'>\n",
                        "Int64Index: 12915 entries, 0 to 12914\n",
                        "Data columns (total 22 columns):\n",
                        "gvkey           12915 non-null int64\n",
                        "gyear           12915 non-null int64\n",
                        "patent_count    12915 non-null float64\n",
                        "year            12915 non-null int64\n",
                        "blkshval        3905 non-null float64\n",
                        "reload          3905 non-null float64\n",
                        "co_per_rol      3905 non-null float64\n",
                        "grntnum         3905 non-null float64\n",
                        "numsecur        3905 non-null float64\n",
                        "pcttotopt       3885 non-null float64\n",
                        "mktpric         3903 non-null float64\n",
                        "expric          3903 non-null float64\n",
                        "pcdvalue        3876 non-null object\n",
                        "value           3870 non-null float64\n",
                        "exdate          3898 non-null float64\n",
                        "pceo            3905 non-null object\n",
                        "cusip           3905 non-null object\n",
                        "sicdesc         3905 non-null object\n",
                        "naicsdesc       3585 non-null object\n",
                        "ticker          3905 non-null object\n",
                        "naics           3585 non-null float64\n",
                        "sic             12915 non-null float64\n",
                        "dtypes: float64(13), int64(3), object(6)\n",
                        "memory usage: 2.3+ MB\n"
                    ]
                }
            ],
            "source": [
                "pat_full.info()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Before going to visualization, we would like to aggregate `sic` into larger industry groups. According to [SIC classification](https://en.wikipedia.org/wiki/Standard_Industrial_Classification), SIC classify companies into 10 major categories, i.e. Agriculture, Forestry and Fishing, Mining, Construction, etc."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 21,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>60.0</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>...</td>\n",
                            "      <td>518.838</td>\n",
                            "      <td>20050117.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>825.424</td>\n",
                            "      <td>20051011.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>2515.578</td>\n",
                            "      <td>20060709.0</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 23 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "0   1004   1992           1.0  1992       NaN     NaN         NaN      NaN   \n",
                            "1   1004   1993           3.0  1993       NaN     NaN         NaN      NaN   \n",
                            "2   1004   1994           1.0  1994   288.811     0.0      5623.0      1.0   \n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...     value      exdate pceo  cusip  \\\n",
                            "0       NaN        NaN  ...       NaN         NaN  NaN    NaN   \n",
                            "1       NaN        NaN  ...       NaN         NaN  NaN    NaN   \n",
                            "2      60.0      24.00  ...   518.838  20050117.0  CEO  36110   \n",
                            "3      75.0      26.05  ...   825.424  20051011.0  CEO  36110   \n",
                            "4     200.0      32.08  ...  2515.578  20060709.0  CEO  36110   \n",
                            "\n",
                            "                        sicdesc  \\\n",
                            "0                           NaN   \n",
                            "1                           NaN   \n",
                            "2  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "3  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                           naicsdesc ticker     naics     sic  \\\n",
                            "0                                                NaN    NaN       NaN  5080.0   \n",
                            "1                                                NaN    NaN       NaN  5080.0   \n",
                            "2  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0   \n",
                            "3  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0   \n",
                            "4  Transportation Equipment and Supplies (except ...    AIR  423860.0  5080.0   \n",
                            "\n",
                            "            sector  \n",
                            "0  Wholesale Trade  \n",
                            "1  Wholesale Trade  \n",
                            "2  Wholesale Trade  \n",
                            "3  Wholesale Trade  \n",
                            "4  Wholesale Trade  \n",
                            "\n",
                            "[5 rows x 23 columns]"
                        ]
                    },
                    "execution_count": 21,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat_full['sector'] = pd.cut(pat_full['sic'], \n",
                "                            bins=[100, 999, 1499, 1999, 3999, 4999, 5199, 5999, 6799, 8999, 9729], \n",
                "                            labels=['Agriculture, Forestry and Fishing', 'Mining', 'Construction',\n",
                "                                    'Manufacturing','Transportation, Communications and Utility',\n",
                "                                    'Wholesale Trade','Retail Trade','Finance, Insurance and Real Estate',\n",
                "                                    'Services','Public Administration'])\n",
                "pat_full.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "As we can see below, there are no observations for companies in Public Administration and Agriculture in this merged datase"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 21,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>0 rows × 23 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "Empty DataFrame\n",
                            "Columns: [gvkey, gyear, patent_count, year, blkshval, reload, co_per_rol, grntnum, numsecur, pcttotopt, mktpric, expric, pcdvalue, value, exdate, pceo, cusip, sicdesc, naicsdesc, ticker, naics, sic, sector]\n",
                            "Index: []\n",
                            "\n",
                            "[0 rows x 23 columns]"
                        ]
                    },
                    "execution_count": 21,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat_full.loc[pat_full['sector']=='Public Administration']"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 22,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>value</th>\n",
                            "      <th>exdate</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>0 rows × 23 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "Empty DataFrame\n",
                            "Columns: [gvkey, gyear, patent_count, year, blkshval, reload, co_per_rol, grntnum, numsecur, pcttotopt, mktpric, expric, pcdvalue, value, exdate, pceo, cusip, sicdesc, naicsdesc, ticker, naics, sic, sector]\n",
                            "Index: []\n",
                            "\n",
                            "[0 rows x 23 columns]"
                        ]
                    },
                    "execution_count": 22,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat_full.loc[pat_full['sector']=='Agriculture, Forestry and Fishing']"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## **Visualiztion**"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "##### Mean Measure of Company Value/Patent Count/Stock Option Value for Each Industry"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 23,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "image/png": "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\n",
                        "text/plain": [
                            "<Figure size 720x720 with 3 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    },
                    "output_type": "display_data"
                }
            ],
            "source": [
                "vsec_mean = pat_full.groupby(['sector'])['value'].mean().sort_values() # mean company value\n",
                "psec_mean = pat_full.groupby(['sector'])['patent_count'].mean().sort_values() # mean patent count\n",
                "stsec_mean = pat_full.groupby(['sector'])['blkshval'].mean().sort_values() # mean CEO's stock option value\n",
                "\n",
                "\n",
                "\n",
                "fig, ax = plt.subplots(3, 1, figsize=(10, 10))\n",
                "ax[0].barh(vsec_mean.index, vsec_mean.values, color='orange')\n",
                "\n",
                "ax[0].set_xlabel(\"Mean Company Value\")\n",
                "ax[0].set_ylabel( \"Industry\")\n",
                "ax[0].set_title(\"Mean Comapny Value Each Industry from 1992-2006\")\n",
                "\n",
                "ax[1].barh(psec_mean.index, psec_mean.values, color='orange')\n",
                "\n",
                "ax[1].set_xlabel(\"Mean Number of Patent Granted\")\n",
                "ax[1].set_ylabel( \"Industry\")\n",
                "ax[1].set_title(\"Mean Number of Patent Granted for Each Industry from 1992-2006\")\n",
                "\n",
                "ax[2].barh(stsec_mean.index, stsec_mean.values, color='orange')\n",
                "\n",
                "ax[2].set_xlabel(\"Mean CEO's Granted Stock Option Value\")\n",
                "ax[2].set_ylabel( \"Industry\")\n",
                "ax[2].set_title(\"Mean CEO's Granted Stock Option Value for Each Industry from 1992-2006\")\n",
                "\n",
                "\n",
                "fig.tight_layout()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Trend in patent data and value of CEO's granted stock option"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 22,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "image/png": "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\n",
                        "text/plain": [
                            "<Figure size 720x432 with 2 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    },
                    "output_type": "display_data"
                }
            ],
            "source": [
                "pc_mean = pgvkey3.groupby([\"gyear\"])[\"patent_count\"].mean()\n",
                "stockog3 = stockog2.groupby([\"year\"])[\"blkshval\"].mean()\n",
                "\n",
                "fig, ax = plt.subplots(1,2, figsize=(10, 6))\n",
                "ax[0].bar(range(pc_mean.shape[0]), pc_mean.values, color=\"blue\")\n",
                "ax[1].bar(range(stockog3.shape[0]), stockog3.values, color=\"blue\")\n",
                "\n",
                "# xtickes labels\n",
                "ax[0].set_xticks(range(pc_mean.shape[0]))\n",
                "ax[0].set_xticklabels(pc_mean.index, rotation=50,fontsize=10)\n",
                "ax[1].set_xticks(range(stockog3.shape[0]))\n",
                "ax[1].set_xticklabels(stockog3.index, rotation=50,fontsize=10)\n",
                "\n",
                "# x-axis label\n",
                "ax[0].set_ylabel(\"Mean Number of Patent Granted\")\n",
                "ax[0].set_xlabel( \"Year\")\n",
                "ax[0].set_title(\"Mean Number of Patent Granted from 1992-2006\")\n",
                "ax[1].set_ylabel(\"Mean Value of CEO's Granted Stock Option\")\n",
                "ax[1].set_xlabel( \"Year\")\n",
                "ax[1].set_title(\"Mean Stock Option Value for CEOs from 1992-2006\")\n",
                "\n",
                "fig.tight_layout()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Regression\n",
                "\n",
                "All patents are granted from United States Patent Office (USPO): the average pendency of a utility non-provisional patent application to final disposal is approximately 25.7 months (this is also evident in the data). I will examine the relationship between patent count for a 3-year lagged measure of CEO's granted stock option value. Selection of 3-year as the time lag seems arbitrary, as a result, I will perform another specification where I use 5-year as the time lag. Furthermore, due to the average pendency of the patent process, it's a natural placebo test to have in this study where I use the same-year CEO's stock option value and we should not expect any significant result in this specification. Summary of specifications is as follows:\n",
                "\n",
                "* Same-year CEO's granted stock option as independent variable (placebo test) - Specification 1\n",
                "* 3-year lagged measure CEO's granted stock option as independent variable (baseline regression) - Specification 2\n",
                "* 5-year lagged measure CEO's granted stock option as independent variable - Specification 3"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Empirical Evidence on Average Pendency of Patent Application Process"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 23,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/plain": [
                            "appyear  gyear\n",
                            "1997     1999     66991\n",
                            "1999     2001     61859\n",
                            "1996     1998     61361\n",
                            "2000     2002     61203\n",
                            "1998     2000     61117\n",
                            "2001     2003     60241\n",
                            "2002     2004     57828\n",
                            "1995     1997     50220\n",
                            "1994     1996     45368\n",
                            "2003     2006     41728\n",
                            "Name: patent, dtype: int64"
                        ]
                    },
                    "execution_count": 23,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pats.groupby([\"appyear\",\"gyear\"]).nunique()[\"patent\"].nlargest(10)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 24,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/plain": [
                            "count    4.176878e+06\n",
                            "mean     2.208581e+00\n",
                            "std      1.197069e+00\n",
                            "min      0.000000e+00\n",
                            "25%      1.000000e+00\n",
                            "50%      2.000000e+00\n",
                            "75%      3.000000e+00\n",
                            "max      3.600000e+01\n",
                            "Name: pendency, dtype: float64"
                        ]
                    },
                    "execution_count": 24,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pats['pendency'] = pats['gyear']-pats['appyear']\n",
                "pats['pendency'].describe()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Create Lagged Measured of CEO's Granted Stock Option"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 25,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "      <th>blkshval3</th>\n",
                            "      <th>blkshval5</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>0</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1992</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1993</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>1994</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>60.0</td>\n",
                            "      <td>24.00</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>150.0</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>195.0</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>7 rows × 25 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "0   1004   1992           1.0  1992       NaN     NaN         NaN      NaN   \n",
                            "1   1004   1993           3.0  1993       NaN     NaN         NaN      NaN   \n",
                            "2   1004   1994           1.0  1994   288.811     0.0      5623.0      1.0   \n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "5   1004   1997           0.0  1997  2528.623     0.0      5623.0      1.0   \n",
                            "6   1004   1998           0.0  1998  1644.472     0.0      5623.0      1.0   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...  pceo  cusip                       sicdesc  \\\n",
                            "0       NaN        NaN  ...   NaN    NaN                           NaN   \n",
                            "1       NaN        NaN  ...   NaN    NaN                           NaN   \n",
                            "2      60.0      24.00  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "3      75.0      26.05  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4     200.0      32.08  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "5     150.0      16.84  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "6     195.0      23.59  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                           naicsdesc  ticker     naics  \\\n",
                            "0                                                NaN     NaN       NaN   \n",
                            "1                                                NaN     NaN       NaN   \n",
                            "2  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "3  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "4  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "5  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "6  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "\n",
                            "      sic           sector blkshval3 blkshval5  \n",
                            "0  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "1  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "2  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "3  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "4  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "5  5080.0  Wholesale Trade   288.811       NaN  \n",
                            "6  5080.0  Wholesale Trade   325.076       NaN  \n",
                            "\n",
                            "[7 rows x 25 columns]"
                        ]
                    },
                    "execution_count": 25,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# shift the blkshval and concat it to our original dataset\n",
                "\n",
                "time = 3 # 3-year lagged mesure\n",
                "\n",
                "blkshval3 = pat_full.groupby([\"gvkey\"])[\"blkshval\"].shift(time).to_frame().rename(columns={\"blkshval\":\"blkshval3\"})\n",
                "patlg_full = pd.concat([pat_full,blkshval3], axis=1)\n",
                "\n",
                "lag = 2 # 5-year lagged measure\n",
                "blkshval5 = pat_full.groupby([\"gvkey\"])[\"blkshval\"].shift(time+lag).to_frame().rename(columns={\"blkshval\":\"blkshval5\"})\n",
                "patlg_full1 = pd.concat([patlg_full,blkshval5], axis=1)\n",
                "\n",
                "patlg_full1.head(7)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Creating lagged measure of CEO's granted stock option value generates additional missing values as earliest data we have is 1992. To increase the quality of my dataset, I will adjust the year range of our merged dataset to 1995-2006 in line with our baseline regression; and filter out companies with more than 5 missing values of blkshval (5 missing years) over the period of 1995-2006 "
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Adjust year range\n",
                "\n",
                "For our baseline regression, where I use 3-year lagged measure of CEO's stock option value, since year range for our the independent variable, CEO stock option is from 1992 to 2006, the earliest year for patent count would be from 1995. Adjusted year range for independent and dependent variables:  \n",
                "* CEO's stock option: 1992-2003 (this is automatically adjusted when shifting the data)\n",
                "* Patent count: 1995-2006\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 26,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "      <th>blkshval3</th>\n",
                            "      <th>blkshval5</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>150.0</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>195.0</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>7</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>288.811</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 25 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "5   1004   1997           0.0  1997  2528.623     0.0      5623.0      1.0   \n",
                            "6   1004   1998           0.0  1998  1644.472     0.0      5623.0      1.0   \n",
                            "7   1004   1999           0.0  1999       NaN     NaN         NaN      NaN   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...  pceo  cusip                       sicdesc  \\\n",
                            "3      75.0      26.05  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4     200.0      32.08  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "5     150.0      16.84  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "6     195.0      23.59  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "7       NaN        NaN  ...   NaN    NaN                           NaN   \n",
                            "\n",
                            "                                           naicsdesc  ticker     naics  \\\n",
                            "3  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "4  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "5  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "6  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "7                                                NaN     NaN       NaN   \n",
                            "\n",
                            "      sic           sector blkshval3 blkshval5  \n",
                            "3  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "4  5080.0  Wholesale Trade       NaN       NaN  \n",
                            "5  5080.0  Wholesale Trade   288.811       NaN  \n",
                            "6  5080.0  Wholesale Trade   325.076       NaN  \n",
                            "7  5080.0  Wholesale Trade  1443.320   288.811  \n",
                            "\n",
                            "[5 rows x 25 columns]"
                        ]
                    },
                    "execution_count": 26,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "patlgg_full1 = patlg_full1.loc[patlg_full[\"gyear\"]>=(1992+time)]\n",
                "patlgg_full1.head()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Filter Company with too many missing values\n",
                "\n",
                "There are too many missing values in `blkshval` measure, we would like to filter companies with too many missing values to reduce the noise  "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 27,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>pceo</th>\n",
                            "      <th>cusip</th>\n",
                            "      <th>sicdesc</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "      <th>blkshval3</th>\n",
                            "      <th>blkshval5</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.000</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.000</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>150.000</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>195.000</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>7</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>288.811</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>8</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>2000</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>2000</td>\n",
                            "      <td>2825.003</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>600.000</td>\n",
                            "      <td>48.53</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>325.076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>9</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>2001</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>2001</td>\n",
                            "      <td>1280.686</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>175.473</td>\n",
                            "      <td>27.55</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>1443.320</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>10</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>2002</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>2002</td>\n",
                            "      <td>1267.923</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>280.000</td>\n",
                            "      <td>29.68</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>2528.623</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>11</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>2003</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>2003</td>\n",
                            "      <td>2668.131</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>356.393</td>\n",
                            "      <td>23.38</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>2825.003</td>\n",
                            "      <td>1644.472</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>12</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>2004</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>2004</td>\n",
                            "      <td>4916.649</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>25.724</td>\n",
                            "      <td>3.04</td>\n",
                            "      <td>...</td>\n",
                            "      <td>CEO</td>\n",
                            "      <td>36110</td>\n",
                            "      <td>MACHINERY AND EQUIPMENT-WHSL</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1280.686</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>10 rows × 25 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "    gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "3    1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4    1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "5    1004   1997           0.0  1997  2528.623     0.0      5623.0      1.0   \n",
                            "6    1004   1998           0.0  1998  1644.472     0.0      5623.0      1.0   \n",
                            "7    1004   1999           0.0  1999       NaN     NaN         NaN      NaN   \n",
                            "8    1004   2000           1.0  2000  2825.003     0.0      5623.0      1.0   \n",
                            "9    1004   2001           0.0  2001  1280.686     0.0      5623.0      1.0   \n",
                            "10   1004   2002           0.0  2002  1267.923     0.0      5623.0      1.0   \n",
                            "11   1004   2003           1.0  2003  2668.131     0.0      5623.0      1.0   \n",
                            "12   1004   2004           0.0  2004  4916.649     1.0      5623.0      1.0   \n",
                            "\n",
                            "    numsecur  pcttotopt  ...  pceo  cusip                       sicdesc  \\\n",
                            "3     75.000      26.05  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "4    200.000      32.08  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "5    150.000      16.84  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "6    195.000      23.59  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "7        NaN        NaN  ...   NaN    NaN                           NaN   \n",
                            "8    600.000      48.53  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "9    175.473      27.55  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "10   280.000      29.68  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "11   356.393      23.38  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "12    25.724       3.04  ...   CEO  36110  MACHINERY AND EQUIPMENT-WHSL   \n",
                            "\n",
                            "                                            naicsdesc  ticker     naics  \\\n",
                            "3   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "4   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "5   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "6   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "7                                                 NaN     NaN       NaN   \n",
                            "8   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "9   Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "10  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "11  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "12  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "\n",
                            "       sic           sector blkshval3 blkshval5  \n",
                            "3   5080.0  Wholesale Trade       NaN       NaN  \n",
                            "4   5080.0  Wholesale Trade       NaN       NaN  \n",
                            "5   5080.0  Wholesale Trade   288.811       NaN  \n",
                            "6   5080.0  Wholesale Trade   325.076       NaN  \n",
                            "7   5080.0  Wholesale Trade  1443.320   288.811  \n",
                            "8   5080.0  Wholesale Trade  2528.623   325.076  \n",
                            "9   5080.0  Wholesale Trade  1644.472  1443.320  \n",
                            "10  5080.0  Wholesale Trade       NaN  2528.623  \n",
                            "11  5080.0  Wholesale Trade  2825.003  1644.472  \n",
                            "12  5080.0  Wholesale Trade  1280.686       NaN  \n",
                            "\n",
                            "[10 rows x 25 columns]"
                        ]
                    },
                    "execution_count": 27,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat35 = patlgg_full1.loc[patlgg_full1.groupby(['gvkey'])['blkshval'].filter(lambda x: len(x[pd.isnull(x)])< 5).index]\n",
                "pat35.head(10)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Log Measure of Stock Option Value"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 28,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>naicsdesc</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "      <th>blkshval3</th>\n",
                            "      <th>blkshval5</th>\n",
                            "      <th>log_blkshval</th>\n",
                            "      <th>log_blkshval3</th>\n",
                            "      <th>log_blkshval5</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>150.0</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>...</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.835430</td>\n",
                            "      <td>5.665772</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>195.0</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>...</td>\n",
                            "      <td>Transportation Equipment and Supplies (except ...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.405175</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>7</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>5.665772</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 28 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "5   1004   1997           0.0  1997  2528.623     0.0      5623.0      1.0   \n",
                            "6   1004   1998           0.0  1998  1644.472     0.0      5623.0      1.0   \n",
                            "7   1004   1999           0.0  1999       NaN     NaN         NaN      NaN   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...  \\\n",
                            "3      75.0      26.05  ...   \n",
                            "4     200.0      32.08  ...   \n",
                            "5     150.0      16.84  ...   \n",
                            "6     195.0      23.59  ...   \n",
                            "7       NaN        NaN  ...   \n",
                            "\n",
                            "                                           naicsdesc  ticker     naics  \\\n",
                            "3  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "4  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "5  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "6  Transportation Equipment and Supplies (except ...     AIR  423860.0   \n",
                            "7                                                NaN     NaN       NaN   \n",
                            "\n",
                            "      sic           sector blkshval3 blkshval5 log_blkshval log_blkshval3  \\\n",
                            "3  5080.0  Wholesale Trade       NaN       NaN     5.784059           NaN   \n",
                            "4  5080.0  Wholesale Trade       NaN       NaN     7.274701           NaN   \n",
                            "5  5080.0  Wholesale Trade   288.811       NaN     7.835430      5.665772   \n",
                            "6  5080.0  Wholesale Trade   325.076       NaN     7.405175      5.784059   \n",
                            "7  5080.0  Wholesale Trade  1443.320   288.811          NaN      7.274701   \n",
                            "\n",
                            "  log_blkshval5  \n",
                            "3           NaN  \n",
                            "4           NaN  \n",
                            "5           NaN  \n",
                            "6           NaN  \n",
                            "7      5.665772  \n",
                            "\n",
                            "[5 rows x 28 columns]"
                        ]
                    },
                    "execution_count": 28,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat35['log_blkshval'] = np.log(pat35['blkshval'].replace(0, np.nan))\n",
                "pat35['log_blkshval3'] = np.log(pat35['blkshval3'].replace(0, np.nan))\n",
                "pat35['log_blkshval5'] = np.log(pat35['blkshval5'].replace(0, np.nan))\n",
                "pat35.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 32,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "<class 'pandas.core.frame.DataFrame'>\n",
                        "Int64Index: 1356 entries, 3 to 11894\n",
                        "Data columns (total 28 columns):\n",
                        "gvkey            1356 non-null int64\n",
                        "gyear            1356 non-null int64\n",
                        "patent_count     1356 non-null float64\n",
                        "year             1356 non-null int64\n",
                        "blkshval         1057 non-null float64\n",
                        "reload           1057 non-null float64\n",
                        "co_per_rol       1057 non-null float64\n",
                        "grntnum          1057 non-null float64\n",
                        "numsecur         1057 non-null float64\n",
                        "pcttotopt        1056 non-null float64\n",
                        "mktpric          1056 non-null float64\n",
                        "expric           1056 non-null float64\n",
                        "pcdvalue         1046 non-null object\n",
                        "value            1043 non-null float64\n",
                        "exdate           1056 non-null float64\n",
                        "pceo             1057 non-null object\n",
                        "cusip            1057 non-null object\n",
                        "sicdesc          1057 non-null object\n",
                        "naicsdesc        1057 non-null object\n",
                        "ticker           1057 non-null object\n",
                        "naics            1057 non-null float64\n",
                        "sic              1356 non-null float64\n",
                        "sector           1344 non-null category\n",
                        "blkshval3        944 non-null float64\n",
                        "blkshval5        742 non-null float64\n",
                        "log_blkshval     1056 non-null float64\n",
                        "log_blkshval3    941 non-null float64\n",
                        "log_blkshval5    739 non-null float64\n",
                        "dtypes: category(1), float64(18), int64(3), object(6)\n",
                        "memory usage: 298.3+ KB\n"
                    ]
                }
            ],
            "source": [
                "pat35.info()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Scatter Plot\n",
                "\n",
                "We are ready to examin the relationship between CEO's granted stock option value and patent count, before running regressions, I would like to have scatter plots to visualize the relationship between stock option value and granted patent count across three specifications"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 33,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "image/png": "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\n",
                        "text/plain": [
                            "<Figure size 720x432 with 3 Axes>"
                        ]
                    },
                    "metadata": {
                        "needs_background": "light"
                    },
                    "output_type": "display_data"
                }
            ],
            "source": [
                "import seaborn as sns\n",
                "\n",
                "fig, ax = plt.subplots(1,3, figsize=(10, 6))\n",
                "\n",
                "sns.regplot(x='log_blkshval', y='patent_count', data=pat35, \n",
                "            scatter_kws={\"color\": \"grey\"}, line_kws={\"color\": \"blue\"},marker='.', ax=ax[0])\n",
                "\n",
                "sns.regplot(x='log_blkshval3', y='patent_count', data=pat35, \n",
                "            scatter_kws={\"color\": \"grey\"}, line_kws={\"color\": \"blue\"},marker='.', ax=ax[1])\n",
                "\n",
                "sns.regplot(x='log_blkshval5', y='patent_count', data=pat35, \n",
                "            scatter_kws={\"color\": \"grey\"}, line_kws={\"color\": \"blue\"}, marker='.', ax=ax[2])\n",
                "\n",
                "ax[0].set_xlabel(\"Same-year Stock Option Value\")\n",
                "ax[1].set_xlabel(\"3-Year Lagged Stock Option Value\")\n",
                "ax[2].set_xlabel(\"5-Year Lagged Stock Option Value\")\n",
                "\n",
                "fig.suptitle('Scatter Plot of Patent Count and CEO\\'s Granted Stock Option Value', fontsize=18)\n",
                "\n",
                "fig.tight_layout(rect=[0, 0.03, 1, 0.95])"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Specification 1 - Placebo Test \n",
                "(same-year stock option value)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 34,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Sample Coefficient for log_blkshval is 1.2010316064564557\n",
                        "P-value for log_blkshval coefficient is 0.1804166169344401\n"
                    ]
                }
            ],
            "source": [
                "from statsmodels.formula.api import ols\n",
                "\n",
                "# generate log measure of company value\n",
                "pat35['log_value'] = np.log(pat35['value'].replace(0, np.nan))\n",
                "\n",
                "# we include company dummies (as firm-level fixed effect) and company value as control variable\n",
                "\n",
                "robust_ols1 = ols('patent_count ~ C(gvkey) + log_blkshval + log_value', data=pat35).fit(cov_type='HC3') \n",
                "# HC3 stands for robust standard error used in this regression\n",
                "\n",
                "\n",
                "print('Sample Coefficient for log_blkshval is', robust_ols1.params[113])\n",
                "print('P-value for log_blkshval coefficient is', robust_ols1.pvalues[113])"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Specification 2 - Baseline Regression\n",
                "(3-year lagged stock option value)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 35,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Sample Coefficient for log_blkshval3 is 1.565737885158954\n",
                        "P-value for log_blkshval3 coefficient is 0.0032442725179993898\n"
                    ]
                }
            ],
            "source": [
                "robust_ols2 = ols('patent_count ~ C(gvkey) + log_blkshval3 + log_value', data=pat35).fit(cov_type='HC3')\n",
                "print('Sample Coefficient for log_blkshval3 is', robust_ols2.params[113])\n",
                "print('P-value for log_blkshval3 coefficient is', robust_ols2.pvalues[113])"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Specification 3 \n",
                "(5-year lagged stock option value)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 36,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Sample Coefficient for log_blkshval5 is 1.3594838259327475\n",
                        "P-value for log_blkshval5 coefficient is 0.018357035129659836\n"
                    ]
                }
            ],
            "source": [
                "robust_ols3 = ols('patent_count ~ C(gvkey) + log_blkshval5 + log_value', data=pat35).fit(cov_type='HC3')\n",
                "print('Sample Coefficient for log_blkshval5 is', robust_ols3.params[113])\n",
                "print('P-value for log_blkshval5 coefficient is', robust_ols3.pvalues[113])"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Interpretation\n",
                "\n",
                "From the regression results, we can see that our placebo test is satisfied in Specification 1; while our results remain robust across two specifications of lagged measure of CEO's granted stock option value. \n",
                "\n",
                "To interpret our result from the baseline regression, we can see that coefficient of 3-year lagged measure of CEO's stock option value is 1.566, that is controlling for company value and firm-fixed effect, every 10% increase in CEO's stock option value is associated with 0.0156 unit increase in granted patent count 3 years after. This result is significant at 1% siginificance level   "
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Prediction (Lasso)\n",
                "\n",
                "I would like to predict company's patent count using baseline specification of CEO's granted stock option value, i.e. 3-year lagged measure of CEO's granted stock option value"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 37,
            "metadata": {},
            "outputs": [],
            "source": [
                "from sklearn import (linear_model, model_selection, metrics)\n",
                "from sklearn.preprocessing import Imputer "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 38,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>gvkey</th>\n",
                            "      <th>gyear</th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>year</th>\n",
                            "      <th>blkshval</th>\n",
                            "      <th>reload</th>\n",
                            "      <th>co_per_rol</th>\n",
                            "      <th>grntnum</th>\n",
                            "      <th>numsecur</th>\n",
                            "      <th>pcttotopt</th>\n",
                            "      <th>...</th>\n",
                            "      <th>ticker</th>\n",
                            "      <th>naics</th>\n",
                            "      <th>sic</th>\n",
                            "      <th>sector</th>\n",
                            "      <th>blkshval3</th>\n",
                            "      <th>blkshval5</th>\n",
                            "      <th>log_blkshval</th>\n",
                            "      <th>log_blkshval3</th>\n",
                            "      <th>log_blkshval5</th>\n",
                            "      <th>log_value</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>3.0</td>\n",
                            "      <td>1995</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>75.0</td>\n",
                            "      <td>26.05</td>\n",
                            "      <td>...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>6.715897</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1996</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>200.0</td>\n",
                            "      <td>32.08</td>\n",
                            "      <td>...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.830258</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1997</td>\n",
                            "      <td>2528.623</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>150.0</td>\n",
                            "      <td>16.84</td>\n",
                            "      <td>...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.835430</td>\n",
                            "      <td>5.665772</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.703844</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1998</td>\n",
                            "      <td>1644.472</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>5623.0</td>\n",
                            "      <td>1.0</td>\n",
                            "      <td>195.0</td>\n",
                            "      <td>23.59</td>\n",
                            "      <td>...</td>\n",
                            "      <td>AIR</td>\n",
                            "      <td>423860.0</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>325.076</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.405175</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.767758</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>7</th>\n",
                            "      <td>1004</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>0.0</td>\n",
                            "      <td>1999</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>...</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>5080.0</td>\n",
                            "      <td>Wholesale Trade</td>\n",
                            "      <td>1443.320</td>\n",
                            "      <td>288.811</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>5.665772</td>\n",
                            "      <td>NaN</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 29 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   gvkey  gyear  patent_count  year  blkshval  reload  co_per_rol  grntnum  \\\n",
                            "3   1004   1995           3.0  1995   325.076     0.0      5623.0      1.0   \n",
                            "4   1004   1996           0.0  1996  1443.320     0.0      5623.0      1.0   \n",
                            "5   1004   1997           0.0  1997  2528.623     0.0      5623.0      1.0   \n",
                            "6   1004   1998           0.0  1998  1644.472     0.0      5623.0      1.0   \n",
                            "7   1004   1999           0.0  1999       NaN     NaN         NaN      NaN   \n",
                            "\n",
                            "   numsecur  pcttotopt  ...  ticker     naics     sic           sector  \\\n",
                            "3      75.0      26.05  ...     AIR  423860.0  5080.0  Wholesale Trade   \n",
                            "4     200.0      32.08  ...     AIR  423860.0  5080.0  Wholesale Trade   \n",
                            "5     150.0      16.84  ...     AIR  423860.0  5080.0  Wholesale Trade   \n",
                            "6     195.0      23.59  ...     AIR  423860.0  5080.0  Wholesale Trade   \n",
                            "7       NaN        NaN  ...     NaN       NaN  5080.0  Wholesale Trade   \n",
                            "\n",
                            "   blkshval3 blkshval5 log_blkshval log_blkshval3 log_blkshval5 log_value  \n",
                            "3        NaN       NaN     5.784059           NaN           NaN  6.715897  \n",
                            "4        NaN       NaN     7.274701           NaN           NaN  7.830258  \n",
                            "5    288.811       NaN     7.835430      5.665772           NaN  7.703844  \n",
                            "6    325.076       NaN     7.405175      5.784059           NaN  7.767758  \n",
                            "7   1443.320   288.811          NaN      7.274701      5.665772       NaN  \n",
                            "\n",
                            "[5 rows x 29 columns]"
                        ]
                    },
                    "execution_count": 38,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "pat35.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 39,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>patent_count</th>\n",
                            "      <th>log_blkshval</th>\n",
                            "      <th>log_blkshval3</th>\n",
                            "      <th>log_blkshval5</th>\n",
                            "      <th>log_value</th>\n",
                            "      <th>1013</th>\n",
                            "      <th>1078</th>\n",
                            "      <th>1281</th>\n",
                            "      <th>1380</th>\n",
                            "      <th>1447</th>\n",
                            "      <th>...</th>\n",
                            "      <th>30990</th>\n",
                            "      <th>31166</th>\n",
                            "      <th>31600</th>\n",
                            "      <th>61355</th>\n",
                            "      <th>61401</th>\n",
                            "      <th>64356</th>\n",
                            "      <th>65142</th>\n",
                            "      <th>65156</th>\n",
                            "      <th>65739</th>\n",
                            "      <th>66032</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>3</th>\n",
                            "      <td>3.0</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>6.715897</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>...</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4</th>\n",
                            "      <td>0.0</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.830258</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>...</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5</th>\n",
                            "      <td>0.0</td>\n",
                            "      <td>7.835430</td>\n",
                            "      <td>5.665772</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.703844</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>...</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>6</th>\n",
                            "      <td>0.0</td>\n",
                            "      <td>7.405175</td>\n",
                            "      <td>5.784059</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.767758</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>...</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>7</th>\n",
                            "      <td>0.0</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>7.274701</td>\n",
                            "      <td>5.665772</td>\n",
                            "      <td>NaN</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>...</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "      <td>0</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>5 rows × 117 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "   patent_count  log_blkshval  log_blkshval3  log_blkshval5  log_value  1013  \\\n",
                            "3           3.0      5.784059            NaN            NaN   6.715897     0   \n",
                            "4           0.0      7.274701            NaN            NaN   7.830258     0   \n",
                            "5           0.0      7.835430       5.665772            NaN   7.703844     0   \n",
                            "6           0.0      7.405175       5.784059            NaN   7.767758     0   \n",
                            "7           0.0           NaN       7.274701       5.665772        NaN     0   \n",
                            "\n",
                            "   1078  1281  1380  1447  ...  30990  31166  31600  61355  61401  64356  \\\n",
                            "3     0     0     0     0  ...      0      0      0      0      0      0   \n",
                            "4     0     0     0     0  ...      0      0      0      0      0      0   \n",
                            "5     0     0     0     0  ...      0      0      0      0      0      0   \n",
                            "6     0     0     0     0  ...      0      0      0      0      0      0   \n",
                            "7     0     0     0     0  ...      0      0      0      0      0      0   \n",
                            "\n",
                            "   65142  65156  65739  66032  \n",
                            "3      0      0      0      0  \n",
                            "4      0      0      0      0  \n",
                            "5      0      0      0      0  \n",
                            "6      0      0      0      0  \n",
                            "7      0      0      0      0  \n",
                            "\n",
                            "[5 rows x 117 columns]"
                        ]
                    },
                    "execution_count": 39,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# keep variables of interest only\n",
                "pat35_ = pat35[['patent_count','log_blkshval', 'log_blkshval3', 'log_blkshval5','log_value']]\n",
                "\n",
                "# Creating dummy variables for gvkey\n",
                "gvkey = pat35['gvkey']\n",
                "company = pd.get_dummies(data=gvkey, drop_first=True) # we drop first dummy to avoid dummy variable trap\n",
                "\n",
                "# add gvkey dummies to the dataframe\n",
                "pat35_dm = pd.concat([pat35_, company], axis=1, join='inner')\n",
                "\n",
                "pat35_dm.head()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 45,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "/opt/conda/lib/python3.6/site-packages/sklearn/utils/deprecation.py:58: DeprecationWarning: Class Imputer is deprecated; Imputer was deprecated in version 0.20 and will be removed in 0.22. Import impute.SimpleImputer from sklearn instead.\n",
                        "  warnings.warn(msg, category=DeprecationWarning)\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>lasso</th>\n",
                            "      <th>linear_reg</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval</th>\n",
                            "      <td>1.470852</td>\n",
                            "      <td>1.199849</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval3</th>\n",
                            "      <td>3.178929</td>\n",
                            "      <td>1.902943</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval5</th>\n",
                            "      <td>2.388448</td>\n",
                            "      <td>2.318076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_value</th>\n",
                            "      <td>2.124837</td>\n",
                            "      <td>0.100011</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1013</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>46.847985</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1078</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>118.506054</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1281</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>6.396107</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1380</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.198670</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1447</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.484581</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1661</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-8.417164</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1981</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>10.410232</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2044</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>28.407469</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2220</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>19.602447</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2255</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>66.600309</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2490</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.979601</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2812</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.241478</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2982</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>8.944308</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3041</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>10.281962</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3157</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>14.147535</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3333</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>4.895399</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3504</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.695021</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3897</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>7.131448</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4078</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.280758</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4371</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>30.540069</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4598</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.978093</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4843</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-6.429366</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4941</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>6.582118</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5020</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>71.637926</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5116</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>27.858381</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5134</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.674014</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>...</th>\n",
                            "      <td>...</td>\n",
                            "      <td>...</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24008</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.649352</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24409</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.138936</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24975</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>18.103018</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25119</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.629118</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25166</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>8.311900</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25341</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>9.763014</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25434</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-6.030736</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25579</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.764666</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28139</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.855931</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28678</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>17.035937</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28758</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.564111</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28917</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>9.950370</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29095</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.894105</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29274</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.774692</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29661</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.822674</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29942</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>4.086253</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30006</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.055335</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30094</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.599238</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30576</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>14.213488</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30865</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.835622</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30990</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.162642</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31166</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.206232</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31600</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.714242</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>61355</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.904121</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>61401</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.104409</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>64356</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>24.111176</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65142</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.870313</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65156</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-3.078932</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65739</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>9.530138</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>66032</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>12.334903</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>116 rows × 2 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "                  lasso  linear_reg\n",
                            "log_blkshval   1.470852    1.199849\n",
                            "log_blkshval3  3.178929    1.902943\n",
                            "log_blkshval5  2.388448    2.318076\n",
                            "log_value      2.124837    0.100011\n",
                            "1013           0.000000   46.847985\n",
                            "1078           0.000000  118.506054\n",
                            "1281          -0.000000    6.396107\n",
                            "1380          -0.000000   -2.198670\n",
                            "1447          -0.000000   -5.484581\n",
                            "1661          -0.000000   -8.417164\n",
                            "1981           0.000000   10.410232\n",
                            "2044           0.000000   28.407469\n",
                            "2220           0.000000   19.602447\n",
                            "2255           0.000000   66.600309\n",
                            "2490          -0.000000   -0.979601\n",
                            "2812          -0.000000    1.241478\n",
                            "2982          -0.000000    8.944308\n",
                            "3041          -0.000000   10.281962\n",
                            "3157           0.000000   14.147535\n",
                            "3333          -0.000000    4.895399\n",
                            "3504          -0.000000    3.695021\n",
                            "3897          -0.000000    7.131448\n",
                            "4078          -0.000000    1.280758\n",
                            "4371           0.000000   30.540069\n",
                            "4598          -0.000000   -5.978093\n",
                            "4843          -0.000000   -6.429366\n",
                            "4941          -0.000000    6.582118\n",
                            "5020           0.000000   71.637926\n",
                            "5116           0.000000   27.858381\n",
                            "5134          -0.000000    3.674014\n",
                            "...                 ...         ...\n",
                            "24008         -0.000000   -1.649352\n",
                            "24409         -0.000000    2.138936\n",
                            "24975          0.000000   18.103018\n",
                            "25119         -0.000000    5.629118\n",
                            "25166         -0.000000    8.311900\n",
                            "25341          0.000000    9.763014\n",
                            "25434         -0.000000   -6.030736\n",
                            "25579         -0.000000    5.764666\n",
                            "28139         -0.000000   -1.855931\n",
                            "28678          0.000000   17.035937\n",
                            "28758         -0.000000    1.564111\n",
                            "28917          0.000000    9.950370\n",
                            "29095         -0.000000   -2.894105\n",
                            "29274         -0.000000    2.774692\n",
                            "29661         -0.000000    3.822674\n",
                            "29942         -0.000000    4.086253\n",
                            "30006         -0.000000    0.055335\n",
                            "30094         -0.000000    0.599238\n",
                            "30576          0.000000   14.213488\n",
                            "30865         -0.000000    0.835622\n",
                            "30990         -0.000000   -5.162642\n",
                            "31166         -0.000000   -2.206232\n",
                            "31600         -0.000000    5.714242\n",
                            "61355         -0.000000   -0.904121\n",
                            "61401         -0.000000   -1.104409\n",
                            "64356          0.000000   24.111176\n",
                            "65142         -0.000000    2.870313\n",
                            "65156         -0.000000   -3.078932\n",
                            "65739         -0.000000    9.530138\n",
                            "66032         -0.000000   12.334903\n",
                            "\n",
                            "[116 rows x 2 columns]"
                        ]
                    },
                    "execution_count": 45,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "# X variable\n",
                "X = pat35_dm.drop(['patent_count'], axis=1).copy()\n",
                "\n",
                "# Y variable\n",
                "y = pat35_dm['patent_count']\n",
                "\n",
                "\n",
                "# Transform the inputs\n",
                "X1 = Imputer().fit_transform(X)\n",
                "\n",
                "lr_model = linear_model.LinearRegression()\n",
                "lasso_model = linear_model.Lasso()\n",
                "\n",
                "lr_model.fit(X1,y)\n",
                "lasso_model.fit(X1, y)\n",
                "\n",
                "lasso_coef = lasso_model.coef_.tolist()\n",
                "lr_coef = lr_model.coef_.tolist()\n",
                "\n",
                "lasso_coefs = pd.Series(dict(zip(list(X.columns), lasso_coef)))\n",
                "lr_coefs = pd.Series(dict(zip(list(X.columns), lr_coef)))\n",
                "coefs = pd.DataFrame(dict(lasso=lasso_coefs, linear_reg=lr_coefs))\n",
                "coefs"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Cross-Validation"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 46,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/html": [
                            "<div>\n",
                            "<style scoped>\n",
                            "    .dataframe tbody tr th:only-of-type {\n",
                            "        vertical-align: middle;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe tbody tr th {\n",
                            "        vertical-align: top;\n",
                            "    }\n",
                            "\n",
                            "    .dataframe thead th {\n",
                            "        text-align: right;\n",
                            "    }\n",
                            "</style>\n",
                            "<table border=\"1\" class=\"dataframe\">\n",
                            "  <thead>\n",
                            "    <tr style=\"text-align: right;\">\n",
                            "      <th></th>\n",
                            "      <th>lasso_op</th>\n",
                            "      <th>lasso</th>\n",
                            "      <th>linreg</th>\n",
                            "    </tr>\n",
                            "  </thead>\n",
                            "  <tbody>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval</th>\n",
                            "      <td>0.713116</td>\n",
                            "      <td>1.470852</td>\n",
                            "      <td>1.199849</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval3</th>\n",
                            "      <td>1.612388</td>\n",
                            "      <td>3.178929</td>\n",
                            "      <td>1.902943</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_blkshval5</th>\n",
                            "      <td>1.901507</td>\n",
                            "      <td>2.388448</td>\n",
                            "      <td>2.318076</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>log_value</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.124837</td>\n",
                            "      <td>0.100011</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1013</th>\n",
                            "      <td>39.821478</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>46.847985</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1078</th>\n",
                            "      <td>112.636617</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>118.506054</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1281</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>6.396107</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1380</th>\n",
                            "      <td>-4.592643</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.198670</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1447</th>\n",
                            "      <td>-6.991575</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.484581</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1661</th>\n",
                            "      <td>-9.304681</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-8.417164</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>1981</th>\n",
                            "      <td>1.133060</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>10.410232</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2044</th>\n",
                            "      <td>20.781175</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>28.407469</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2220</th>\n",
                            "      <td>10.616264</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>19.602447</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2255</th>\n",
                            "      <td>60.737519</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>66.600309</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2490</th>\n",
                            "      <td>-3.637032</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.979601</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2812</th>\n",
                            "      <td>-2.005935</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.241478</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>2982</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>8.944308</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3041</th>\n",
                            "      <td>2.166294</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>10.281962</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3157</th>\n",
                            "      <td>6.824688</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>14.147535</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3333</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>4.895399</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3504</th>\n",
                            "      <td>-0.027882</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.695021</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>3897</th>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>7.131448</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4078</th>\n",
                            "      <td>-1.898321</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.280758</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4371</th>\n",
                            "      <td>23.704513</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>30.540069</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4598</th>\n",
                            "      <td>-7.448862</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.978093</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4843</th>\n",
                            "      <td>-7.849594</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-6.429366</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>4941</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>6.582118</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5020</th>\n",
                            "      <td>66.548854</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>71.637926</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5116</th>\n",
                            "      <td>20.020494</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>27.858381</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>5134</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.674014</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>...</th>\n",
                            "      <td>...</td>\n",
                            "      <td>...</td>\n",
                            "      <td>...</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24008</th>\n",
                            "      <td>-3.291582</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.649352</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24409</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.138936</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>24975</th>\n",
                            "      <td>12.111272</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>18.103018</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25119</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.629118</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25166</th>\n",
                            "      <td>1.293048</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>8.311900</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25341</th>\n",
                            "      <td>0.722784</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>9.763014</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25434</th>\n",
                            "      <td>-7.334639</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-6.030736</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>25579</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.764666</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28139</th>\n",
                            "      <td>-4.096579</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.855931</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28678</th>\n",
                            "      <td>10.137569</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>17.035937</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28758</th>\n",
                            "      <td>-1.786547</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>1.564111</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>28917</th>\n",
                            "      <td>0.636853</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>9.950370</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29095</th>\n",
                            "      <td>-4.632701</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.894105</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29274</th>\n",
                            "      <td>-0.670487</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.774692</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29661</th>\n",
                            "      <td>-0.118406</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>3.822674</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>29942</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>4.086253</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30006</th>\n",
                            "      <td>-3.014096</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.055335</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30094</th>\n",
                            "      <td>-2.517042</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.599238</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30576</th>\n",
                            "      <td>7.123058</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>14.213488</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30865</th>\n",
                            "      <td>-2.299679</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>0.835622</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>30990</th>\n",
                            "      <td>-6.486867</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-5.162642</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31166</th>\n",
                            "      <td>-4.352927</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-2.206232</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>31600</th>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>5.714242</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>61355</th>\n",
                            "      <td>-3.368846</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-0.904121</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>61401</th>\n",
                            "      <td>-3.678402</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-1.104409</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>64356</th>\n",
                            "      <td>18.026775</td>\n",
                            "      <td>0.000000</td>\n",
                            "      <td>24.111176</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65142</th>\n",
                            "      <td>-0.106756</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>2.870313</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65156</th>\n",
                            "      <td>-5.245606</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>-3.078932</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>65739</th>\n",
                            "      <td>2.190080</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>9.530138</td>\n",
                            "    </tr>\n",
                            "    <tr>\n",
                            "      <th>66032</th>\n",
                            "      <td>5.376386</td>\n",
                            "      <td>-0.000000</td>\n",
                            "      <td>12.334903</td>\n",
                            "    </tr>\n",
                            "  </tbody>\n",
                            "</table>\n",
                            "<p>116 rows × 3 columns</p>\n",
                            "</div>"
                        ],
                        "text/plain": [
                            "                 lasso_op     lasso      linreg\n",
                            "log_blkshval     0.713116  1.470852    1.199849\n",
                            "log_blkshval3    1.612388  3.178929    1.902943\n",
                            "log_blkshval5    1.901507  2.388448    2.318076\n",
                            "log_value       -0.000000  2.124837    0.100011\n",
                            "1013            39.821478  0.000000   46.847985\n",
                            "1078           112.636617  0.000000  118.506054\n",
                            "1281             0.000000 -0.000000    6.396107\n",
                            "1380            -4.592643 -0.000000   -2.198670\n",
                            "1447            -6.991575 -0.000000   -5.484581\n",
                            "1661            -9.304681 -0.000000   -8.417164\n",
                            "1981             1.133060  0.000000   10.410232\n",
                            "2044            20.781175  0.000000   28.407469\n",
                            "2220            10.616264  0.000000   19.602447\n",
                            "2255            60.737519  0.000000   66.600309\n",
                            "2490            -3.637032 -0.000000   -0.979601\n",
                            "2812            -2.005935 -0.000000    1.241478\n",
                            "2982             0.000000 -0.000000    8.944308\n",
                            "3041             2.166294 -0.000000   10.281962\n",
                            "3157             6.824688  0.000000   14.147535\n",
                            "3333            -0.000000 -0.000000    4.895399\n",
                            "3504            -0.027882 -0.000000    3.695021\n",
                            "3897             0.000000 -0.000000    7.131448\n",
                            "4078            -1.898321 -0.000000    1.280758\n",
                            "4371            23.704513  0.000000   30.540069\n",
                            "4598            -7.448862 -0.000000   -5.978093\n",
                            "4843            -7.849594 -0.000000   -6.429366\n",
                            "4941            -0.000000 -0.000000    6.582118\n",
                            "5020            66.548854  0.000000   71.637926\n",
                            "5116            20.020494  0.000000   27.858381\n",
                            "5134            -0.000000 -0.000000    3.674014\n",
                            "...                   ...       ...         ...\n",
                            "24008           -3.291582 -0.000000   -1.649352\n",
                            "24409           -0.000000 -0.000000    2.138936\n",
                            "24975           12.111272  0.000000   18.103018\n",
                            "25119           -0.000000 -0.000000    5.629118\n",
                            "25166            1.293048 -0.000000    8.311900\n",
                            "25341            0.722784  0.000000    9.763014\n",
                            "25434           -7.334639 -0.000000   -6.030736\n",
                            "25579           -0.000000 -0.000000    5.764666\n",
                            "28139           -4.096579 -0.000000   -1.855931\n",
                            "28678           10.137569  0.000000   17.035937\n",
                            "28758           -1.786547 -0.000000    1.564111\n",
                            "28917            0.636853  0.000000    9.950370\n",
                            "29095           -4.632701 -0.000000   -2.894105\n",
                            "29274           -0.670487 -0.000000    2.774692\n",
                            "29661           -0.118406 -0.000000    3.822674\n",
                            "29942           -0.000000 -0.000000    4.086253\n",
                            "30006           -3.014096 -0.000000    0.055335\n",
                            "30094           -2.517042 -0.000000    0.599238\n",
                            "30576            7.123058  0.000000   14.213488\n",
                            "30865           -2.299679 -0.000000    0.835622\n",
                            "30990           -6.486867 -0.000000   -5.162642\n",
                            "31166           -4.352927 -0.000000   -2.206232\n",
                            "31600           -0.000000 -0.000000    5.714242\n",
                            "61355           -3.368846 -0.000000   -0.904121\n",
                            "61401           -3.678402 -0.000000   -1.104409\n",
                            "64356           18.026775  0.000000   24.111176\n",
                            "65142           -0.106756 -0.000000    2.870313\n",
                            "65156           -5.245606 -0.000000   -3.078932\n",
                            "65739            2.190080 -0.000000    9.530138\n",
                            "66032            5.376386 -0.000000   12.334903\n",
                            "\n",
                            "[116 rows x 3 columns]"
                        ]
                    },
                    "execution_count": 46,
                    "metadata": {},
                    "output_type": "execute_result"
                }
            ],
            "source": [
                "\n",
                "X_train, X_test, y_train, y_test = model_selection.train_test_split(X1, y, test_size=0.2)\n",
                "\n",
                "lasso = linear_model.LassoCV(cv=5).fit(X_train,y_train)\n",
                "\n",
                "lasso_model_op = linear_model.Lasso(alpha=lasso.alpha_)\n",
                "lasso_model_op.fit(X1, y)\n",
                "\n",
                "lasso_op_coef = lasso_model_op.coef_.tolist()\n",
                "\n",
                "lasso_op_coefs = pd.Series(dict(zip(list(X.columns), lasso_op_coef)))\n",
                "lasso_coefs = pd.Series(dict(zip(list(X.columns), lasso_coef)))\n",
                "lr_coefs = pd.Series(dict(zip(list(X.columns), lr_coef)))\n",
                "\n",
                "coefs = pd.DataFrame(dict(lasso_op=lasso_op_coefs, lasso=lasso_coefs, linreg=lr_coefs))\n",
                "coefs"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "#### Report Mean Square Error"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 47,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "mse for lasso with optimal alpha is 242.4160769514998\n",
                        "mse for lasso regression is 1001.1142539255462\n",
                        "mse for linear regression is 238.43091065398735\n"
                    ]
                }
            ],
            "source": [
                "# lasso regression with optimal alpha\n",
                "mse_lasso_op = metrics.mean_squared_error(y, lasso_model_op.predict(X1))\n",
                "print('mse for lasso with optimal alpha is', mse_lasso_op)\n",
                "\n",
                "# lasso regression\n",
                "mse_lasso = metrics.mean_squared_error(y, lasso_model.predict(X1))\n",
                "print('mse for lasso regression is', mse_lasso)\n",
                "\n",
                "# linear regression\n",
                "mse_linear = metrics.mean_squared_error(y, lr_model.predict(X1))\n",
                "print('mse for linear regression is', mse_linear)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Conclusion\n",
                "\n",
                "From our prediction results, we can see that it seems that our features do not posses large predictive power in predicting number of patent counted for a company (other classification were tested, however, mean squared error remain to be large). Nonetheless, we find robust and significant correlation between CEO's granted stock option value and number of granted patents for a company. Findings in this study offer the empirical evidence in support of the theory in coporate finance literature: the \n",
                "compensation scheme that rewards for the long-term success and indicates the commitment to a long-term compensation plan to CEOs, are essential to motivating innovation at the firm level. \n"
            ]
        }
    ],
    "metadata": {
        "kernelspec": {
            "display_name": "Python 3",
            "language": "python",
            "name": "python3"
        },
        "language_info": {
            "codemirror_mode": {
                "name": "ipython",
                "version": 3
            },
            "file_extension": ".py",
            "mimetype": "text/x-python",
            "name": "python",
            "nbconvert_exporter": "python",
            "pygments_lexer": "ipython3",
            "version": "3.7.3"
        }
    },
    "nbformat": 4,
    "nbformat_minor": 2
}