ipynb/chase_coleman-wald_friedman.ipynb

{
    "cells": [
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## A Problem that stumped Milton Friedman\n",
                "\n",
                "(and that Abraham Wald solved by inventing sequential analysis)\n",
                "\n",
                "#### By [Chase Coleman](https://github.com/cc7768) and [Thomas J. Sargent](http://www.tomsargent.com/)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "We begin by importing some packages called by the code that we will be using in this notebook."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 1,
            "metadata": {
                "collapsed": false
            },
            "outputs": [],
            "source": [
                "import numpy as np\n",
                "import matplotlib.pyplot as plt\n",
                "import scipy.interpolate as interp\n",
                "import scipy.stats as st\n",
                "import seaborn as sb\n",
                "import quantecon as qe\n",
                "from ipywidgets import interact, widgets\n",
                "\n",
                "%matplotlib inline\n",
                "import warnings\n",
                "warnings.filterwarnings('ignore')"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 2,
            "metadata": {
                "collapsed": false
            },
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "A file named Wald_Friedman_utils.py already exists in the specified directory ... skipping download.\n"
                    ]
                }
            ],
            "source": [
                "#-Download supporting Wald_Friedman_utils.py file from GitHub-#\n",
                "qe.fetch_nb_dependencies([\"Wald_Friedman_utils.py\"])\n",
                "from Wald_Friedman_utils import *"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Sequential analysis\n",
                "\n",
                "Key ideas in  play are:\n",
                "\n",
                " * Bayes' Law\n",
                " \n",
                " * Dynamic programming\n",
                "\n",
                " * type I and type II statistical errors\n",
                "    \n",
                "    * a type I error occurs when you reject a  null hypothesis that is true\n",
                "    \n",
                "    * a type II error is when you accept a null hypothesis that is false \n",
                " \n",
                "\n",
                " * Abraham Wald's **sequential probability ratio test**\n",
                " \n",
                " * The **power** of a statistical test\n",
                " \n",
                " * The **critical region** of a statistical test\n",
                " \n",
                " * A **uniformly most powerful test**\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "On pages 137-139 of his book **Two Lucky People** with Rose Friedman, Milton Friedman described a problem presented to him and Allen Wallis during World War II when they worked at the U.S. government's Statistical Research Group at Columbia University.  \n",
                "\n",
                "Let's listen to Milton Friedman tell what happened.\n",
                "\n",
                "   \"In order to understand the story, it is necessary to have an idea of a simple statistical problem, and of the\n",
                "   standard procedure for dealing with it.  The actual problem out of which sequential analysis grew will serve.\n",
                "   The Navy has two alternative designs (say A and B) for a projectile.  It wants to determine which is superior. \n",
                "   To do so it undertakes a series of paired firings. On each round it assigns the value 1 or 0 to A accordingly as\n",
                "   its performance is superior or inferio to that of B and conversely 0 or 1 to B.  The Navy asks the statistician \n",
                "   how to conduct the test and how to analyze the results. \n",
                "   \n",
                "   \"The standard statistical answer was to specify a number of firings (say 1,000) and a pair of percentages\n",
                "   (e.g., 53% and 47%) and tell the client that if A receives a 1in more than 53% of the firings, it can be regarded\n",
                "   as superior; if it receives a 1 in fewer than 47%, B can be regarded as superior; if the percentage is between\n",
                "   47% and 53%, neither can be so regarded.\n",
                "   \n",
                "   \"When Allen Wallis was discussing such a problem with (Navy) Captain Garret L. Schyler, the captain objected that     such a test, to quote from Allen's account, may prove wasteful.  If a wise and seasoned ordnance officer like   Schyler were on the premises, he would see after the first few thousand or even few hundred [rounds] that the   experiment need not be completed either because he new method is obviously inferior or because it is obviously   superior beyond what was hoped for $\\ldots$ \"\n",
                "  \n",
                "  Friedman and Wallis struggled with the problem but after realizing that they were not able to solve it themselves told Abraham Wald it.  That started  Wald on the path that led  *Sequential Analysis*. We'll formulate the problem using dynamic programming.\n",
                "\n",
                "This started Wald on the path that led him to _Sequential Analysis_"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Dynamic programming formulation\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "The following presentation of the problem closely follows Dmitri Berskekas's treatment in **Dynamic Programming and Stochastic Control**.  \n",
                "\n",
                "An i.i.d.  random variable $z$ can take on values \n",
                "\n",
                "  * $z \\in [ v_1,  v_2, \\ldots,  v_n]$ when $z$ is a discrete-valued random variable\n",
                "  \n",
                "  * $ z \\in V$ when $z$ is a continuous random variable.  \n",
                "\n",
                "A decision maker wants to  know which of two probability distributions governs  $z$. To formalize this idea,\n",
                "let $x \\in [x_0, x_1]$ be a hidden state that indexes the two distributions:\n",
                "\n",
                "$$ P(v_k \\mid x) = \\begin{cases} f_0(v_k) & \\mbox{if } x = x_0, \\\\\n",
                "                              f_1(v_k) & \\mbox{if } x = x_1. \\end{cases} $$\n",
                "                              \n",
                "when $z$ is a discrete random variable and a density\n",
                "\n",
                "\n",
                "$$ P(v \\mid x) = \\begin{cases} f_0(v) & \\mbox{if } x = x_0, \\\\\n",
                "                              f_1(v) & \\mbox{if } x = x_1. \\end{cases} $$\n",
                "                              \n",
                "when $v$ is continuously distributed.                              \n",
                "\n",
                "                              \n",
                "\n",
                "Before observing any outcomes, a decision maker believes that the probability that $x = x_0$ is $p_{-1}\\in (0,1)$: \n",
                "\n",
                "$$ p_{-1} = \\textrm{Prob}(x=x_0 \\mid \\textrm{ no observations}) $$\n",
                "\n",
                "After observing $k+1$ observations $z_k, z_{k-1}, \\ldots, z_0$ he believes that the probability that the distribution is $f_0$ is\n",
                "\n",
                "$$ p_k = {\\rm Prob} ( x = x_0 \\mid z_k, z_{k-1}, \\ldots, z_0) $$\n",
                "\n",
                "We can compute this $p_k$ recursively by applying Bayes' law:\n",
                "\n",
                "$$ p_0 = \\frac{ p_{-1} f_0(z_0)}{ p_{-1} f_0(z_0) + (1-p_{-1}) f_1(z_0) } $$\n",
                "\n",
                "and then\n",
                "\n",
                "$$ p_{k+1} = \\frac{ p_k f_0(z_{k+1})}{ p_k f_0(z_{k+1}) + (1-p_k) f_1 (z_{k+1}) }. $$\n",
                "\n",
                "\n",
                "After observing $z_k, z_{k-1}, \\ldots, z_0$, the decision maker believes that $z_{k+1}$ \n",
                "has probability distribution\n",
                "\n",
                "$$ p(z_{k+1}) = p_k f_0(z_{k+1}) + (1-p_k) f_1 (z_{k+1}). $$\n",
                "\n",
                "This is evidently a mixture of distributions $f_0$ and $f_1$, with the weight on $f_0$ being the posterior probability $f_0$ that the distribution is $f_0$. \n",
                "\n",
                "**Remark:** *Because the decision maker believes that $z_{k+1}$ is drawn from a mixture of two i.i.d. distributions, he does *not* believe that the sequence  $[z_{k+1}, z_{k+2}, \\ldots] $ is i.i.d.  Instead, he believes that it is *exchangeable*.  See David Kreps\n",
                "*Notes on the Theory of Choice*, chapter 11, for a discussion.*"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Let's look at some examples of two distributions. Here we'll display two beta distributions.  First, we'll show the two distributions, then we'll show mixtures of these same two distributions with various mixing probabilities $p_k$."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 3,
            "metadata": {
                "collapsed": false
            },
            "outputs": [],
            "source": [
                "# Create two distributions over 50 values for k\n",
                "# We are using a discretized beta distribution\n",
                "p_m1 = np.linspace(0, 1, 50)\n",
                "f0 = np.clip(st.beta.pdf(p_m1, a=1, b=1), 1e-8, np.inf)\n",
                "f0 = f0 / np.sum(f0)\n",
                "f1 = np.clip(st.beta.pdf(p_m1, a=9, b=9), 1e-8, np.inf)\n",
                "f1 = f1 / np.sum(f1)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 4,
            "metadata": {
                "collapsed": false
            },
            "outputs": [
                {
                    "data": {
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAAI7CAYAAADxrW3BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xl8VFWe//9XVRaSkAAhCwlLWJJwwiqyCwgKKIiAoKC2\na7etvdl798xvpn8z0z3fsXu+0z3jdDvdra3d2u4oqICICIiCbMpO2A6EJWACSQh7ErJVff+oojvN\nCKQgVbdS9X4+HjySuvdWnXfgUPnk5JxzXV6vFxERERER8XE7HUBEREREJJyoQBYRERERaUIFsoiI\niIhIEyqQRURERESaUIEsIiIiItKECmQRERERkSZinQ4gIhIOjDHfA76K733RDXwE/LO19vglrl8G\n/J21dutlXvMXwCFr7bNXmemnQBdr7dcuOj4OWArs9+eNA5YDP7XWlgaQ71Fr7R8vce5F4E1gB1Bk\nrY0LMPtwoNpau8MY8ziQaa39aSCvISLiFJf2QRaRaOcvZMcDM621R40xbuAXwFRgiLW21qFclyuQ\nn7PW9vY/jgf+AV+BP9haW9mM144BjltrU69wXXdgn7U2PsDsTwOrrbWvBvI8EZFwoBFkEYlqxphU\n4HvAQGvtUQBrrQf4B2PMBOBB4I/GmIPA88B9wK3AKuB+a+1aY8xP/K9xCPgz8PfW2p7GmBfwFZe/\n8D//3/EVsV2B1621P/ZneBT4IRADHAUetNYeae7XYK2tA/6PMSYf+AHwT/727gc+BZ4BbsQ3Mr4d\n+AowH2hvjNkFTAFeANYAM/0Z/x14zn/MZYz5AfAY0Ab4gbV2ob+A72qtfcz/dfzU/7VtBB4Cphlj\nMoD2F64zxnTzv24PoA74lbX2ZX8hvs7f7mNAKvBDa+1cY0xn4CUgy9/+HGvtPzf370dEJFCagywi\n0W4kUGyt3f8F594FxjV53MVa26dp8WqM6Qv8HTAAXxF6N3CpX83daK0dAQwFvmOM6ewvIJ8CJlhr\nDb5pE1db/C0Ebr7o2CSgp7W2wD/ivBPf1/wI0GCt7WutPeS/drC1tp+1dv1FrxEDuKy1fYGvA8/5\nR6D5gq/Va639A/AZvikev77oumeBFdbaAnwj9E8ZY3L859L9mQbiK/Sf8B//PrDSWtsf399zL2NM\np+b+pYiIBEoFsohEu45AxSXOlfnPX7DoC64ZC3xkrS33j+Q+f5m2XgPwj1SXAd2stRVAuwuj18An\nQK8A8jd1Bt9obVMVQB9jzExjTJK19qfW2mWXeP7iSxz34hvBxVq7HN9vH3ObkcfV9IExJha4BXja\n/1qH8c31Hu+/JAbfCDzAZuBC4VwOTDLGjAbqrLX3W2vLmtG+iMhVUYEsItHuOND5Euc64SvOLjjx\nBdekXnS85DJtnW7yeSMQ45/v/IQxZqcxZjfwc67+vbkHf5sXa+0G4Dv+P8eMMa8YY9pd4vlf9PVd\neJ2mixXP4Pu6A5Xmf62zTY6dBDL9nzdaa2sufI6vYAZ4Et/o+O/xfQ0/u4q2RUSaTQWyiES7dUBH\nY8yALzg3Fd/uEJdzBkhu8vhSxfal3ONvZ4y1tg9wLTs9zAI+uPigtfZta+14fCOybfFNCQmIf672\nBRd+KGhaxF44fjnHAa8xpukodxq+0fRLstZ6rLW/tNZeB4wCHvDPDxcRCQoVyCIS1ay1Z/DtWPGy\nMaYH+HZ4MMb8O773yDeu8BKfATcbYzoaY9rgW5wWiEx8W8GdNMak4ZvDnHyF50CT6QvGmDhjzL/h\nG0H+XdOLjDFfNsb8E4C19hSwB9+UiXp8I9htm9nW/f7XuwU4h2+u9FGgvzHGZYxJx7fY74J6oEPT\nF7HWNgJL8M1jxhiTi2/e9oUfQv5mSkaTr+EZY8xE/8ODwDEuPc9bROSaqUAWkahnrf0vfIvH3vXv\n6rATX3E30Vrb4L/sfy1G8z93A/AisBVfobfwC6695POB14F0Y8xe4FXg/we6GWN+dYnXuaCbMWaX\nMWYPvt0zcvAtAjx30esvAIYYY6wxZifQB9+UhaPAauCwMWbkZfKBryCOMcYU4pvm8FX/Th9zgSqg\nyP938GaT57wD/Icx5j8veq1v4vuBYjfwlv+1LkxLudTX+wzwc/+/zQ5gjbV2xWX+bkRErknI90E2\nxjyJbwW1B/i+tXZjk3MT8c2/awAWW2t/box5BN82S158owtDrLWXmj8nIuIoY8wU4N+stUOcziIi\nIlcnpPsgG2PGAnnW2lHGmAJ8q71HNbnkN/hWOB8FVhpj3rLWPu+/7sLzZ4cys4jI5finFuwBBgNH\n8E2RWOdoKBERuSahnmIxAd/m9Fhr9wAdjDHJAMaYnkCltbbUWuvFt93QxYsw/gX4txDmFRG5LP/u\nDj8BPsRXKKcCP3Myk4iIXJtQ30kvC98dli447j9W5P/YdC/ScprsBWqMGQocttb+zRZGIiJOs9Y+\ni28Os4iIRACnF+l94YrlS5x7lL9uIC8iIiIiEhShHkEuxTdSfEFnfPONL5zLbnKui//YBTcB325O\nI16v1+tyXa72FhEREZEoElBhGOoCeSm+uXnPGWMGAyXW2ioAa22xMSbFGJODrzCeCtwHYIzJBs42\n2W7pslwuFxUVZ698oUSsjIwU9QFRPxD1AVEfEMDXDwIR0gLZWrvOGLPJGLMG3x2YHjfGPAycstYu\nwLc/5hx8W7q9bq0t8j81m4tunyoiIiIiEgwh3wc5RLz6aTG6acRAQP1A1AdEfUB8MjJSAppi4fQi\nPRERERGRsBLqOcgiInIV6hrrufydp/+3GFcMMe6Y4AQSEYlgKpBFRMJUo6eRbcd3suLwJxw8Uxzw\n8xNi2nBD52Hc1HUM6Ykdg5BQRCQyqUAWEQkzNQ3nWVf6GR9/vobK8ycByG3fgzaxbQJ6nZKzpXx0\nZDUfH1nDoIz+jM8ZS6/23YMRWUQkoqhAFhEJE5U1J/n489WsLf2M8421xLnjuLHLDdzcdTSd2mYG\n/HoNngY2l2/nw8Or2FJRyJaKQnq2y2F8zliuS++n6RciIpegAllExGEHTx9mxZFVbK3YgcfroV18\nCrd0v5kxXUaQHNf2ql831h3L8KzBDOt0PftOHWDFkVUUHt/Nn3a8QseEVG7uOpobOg8nMTahBb8a\nEZHWTwWyiIgDPF4P2yp2suLIKg6c9s0v7pKczYRuYxnc6Tri3C339uxyueidmkvv1FzKqsr56PM1\nrD+6kbeKFvHewWWM6jycm7qOIS0xtcXaFBFpzbQPskQk7XspEJ79oK6xnjWln/LRkdVUnj8BQP+0\nAsZ3G0vv1FxcroC26rxq5+qrWF3yKas+X8PpurO4cHF95gAm5oyje7tuIckQCuHYByS01Acix+bN\nG1m9eiXXXz+EG2+8KaDnBroPskaQRURC5HTtGZ7Z/gKHz5YQ545lTJeR3Nx1DFlXMb/4WiXHtWVy\nj/FMzBnLprJtfHhkFZvLt7OlvJC7e9/B2K6jQp5JRORy5s6dw8MPP0KHDsH/bZcKZBGRECg5d5Sn\nt73AydpTjMgawp15U0mOv/r5xS0l1h3LiOwhDM8azO4Te3lx1xze2DufippKZubdjtul+0mJSHio\nq6ujoKBvSNrSO5+ISJDtqrQ8uen3nKw9xbRek3mwz91hURw35XK56Jtm+Luh36FTUiYrjnzCc4Uv\nU9tY53Q0ERHmzHmFurpaVq9eGZL2NIIsIhJEq0vW88be+bhdbh7pdx9DOg1yOtJlpSd25MdDvsVz\nO15h+/Gd/Hrz03xj4Fdo36ad09FExEFvrihiw57yFn3NYQWZ3D0+r1nXGtMHj8dLWlo6GzZ8SknJ\n58yYcVeL5mlKI8giIkHg8Xp4p+g9XrdvkxSbyPeu/1rYF8cXJMUl8fh1jzAyeyiHz5bwq42/peTc\nUadjiUgUO3jwALm5eaxYsZxhw0ZQX19PWdmxoLWnEWQRkRZW11jHi7veYGtFIZ2SMvjmwEfISEpz\nOlZAYt2xPFAwm4zEdN49sIQnN/2er/Z/gL5pxuloIuKAu8fnNXu0NxgOHNjPjTeOY/XqVQAkJiZy\n8uQJOnXKCkp7GkEWEWlBZ+rO8ustf2BrRSH5HXrxoyGPt7ri+AKXy8XkHuN5pN99NHgbeXr7C6wu\nWe90LBGJQpWVx8nIyCQ5ORmAc+fOkpraMWjtaQRZRKSFHK0q4+ltz1N5/iQjsoZwX8FdxLbgDT+c\nMqTTIFITOvCH7S/yun2bippK7si9TTtciEjQrVy5goaGBjIyMgAYOXI0mzdvxO2OCdroMahAFhFp\nEXtO7OOPO16mpuE8U3veyuQeE0J2049Q6NW+Bz8e8m2e3v48yw+v5HjNCR7uew/xMfFORxORCBYT\nE0tJSQmzZt0DwHXX+dZyDB48NKjtqkAWEblGa0s38Lp9CzcuHu57L8OzBjsdKSgyktL40ZDHea7w\nJbZWFHJyyym+MfDLtItPcTqaiESoMWPGOtKufj8mInKVPF4PC/a/z6t75pIYk8B3rv9axBbHF7SN\nS+Lbgx5lRNYQis8c4Vcbf8vRqjKnY4mItCgVyCIiV2l+0WKWFn9ERmIaPx76OHkdejodKSRi3bE8\n2Odupva8lRPnT/Lfm56morrS6VgiIi1GBbKIyFVYd3QjHx5ZRaekTH485NtkJmU4HSmkXC4Xt/Wc\nyL3mTqoaqnlm+wvUNJx3OpaISItQgSwiEqADpw8xZ89bJMUm8o2BXw6720aH0o1dRnJz1zEcqy7n\nzztfw+P1OB1JROSaqUAWEQnAyfOneLbwJTx4+Wr/B8hMSnc6kuNm5t1OQWo+Oyr3sHD/EqfjiIhc\ns5DvYmGMeRIYCXiA71trNzY5NxH4OdAAvG+tfcJ//H7g74B64F+ste+HOreISG1jHX/Y/mfO1p1j\ndv4dFHTMdzpSWIhxx/DV/vfzq42/Zdnhj+mcnBXxixVFJLKFdATZGDMWyLPWjgIeBZ666JLfADOB\nMcCtxpgCY0xH4F+AUcBU4I4QRhYRAcDr9fLy7jc5cq6U0Z2HM67rKKcjhZWkuCS+PvDLJMYm8Oqe\neRw6c9jpSCIiVy3UUywmAPMBrLV7gA7GmGQAY0xPoNJaW2qt9QKL/ddPBJZZa6uttWXW2m+EOLOI\nCEsOfciW8u3ktu/J3b1nRNRNQFpKVttMvtLvfho9jTy7/UVO1Z52OpKIyFUJdYGcBVQ0eXzcf+yL\nzpUDnYEeQFtjzAJjzEpjzPhQBBURuWBreSGLDi6lY0Iqjw14MCJuHx0s/dIMM/Nu53TdWZ7d/hJ1\njfVORxKRCLF580aeeuq/+OSTj4PeltOL9C43BOMCvP6PHYEZwFeAF0KQS0QEgM/PlvLirjnEx8Tz\njYFfJiU+2elIYW98txsZmTWU4rNHeHXPXLxer9ORRCQCzJ07h1tvvY38fBP0tkI9DFLKX0eMwTdC\nfLTJuewm57r4j1UBa/3TLg4YY84aY9Kttccv11BGhm59Gu3UBwSurR+cPn+G59a/RJ2nnh+P/jqD\nuvZuwWSR7TtpD3HioxNsLNtK7049mNFnkmNZ9F4g6gORopEbbxwRkpZCXSAvBX4GPGeMGQyUWGur\nAKy1xcaYFGNMDr7CeCpwH1ANvGCM+SW+keS2VyqOASoqzgbpS5DWICMjRX1ArqkfNHgaeGrLcxyv\nPsHUnrfSs02u+lSAvlxwP7/c+BSvb19AOzowIL1vyDPovUDUByLDnDmvcO5cNe+8s4gxY8YF/PxA\nf0gKaYFsrV1njNlkjFkDNAKPG2MeBk5ZaxcA3wTm4Jta8bq1tgjAGDMPWO8//u1QZhaR6OP1ennD\nzmf/6YMMzhzI5B4TnI7UKrVvk8LXBz7Mk5ue5oWdr/HjId+mc3LWlZ8oImHn7aJFbCkvbNHXvD5z\nAHfmTW3Wtcb0wePxMmbMOFavXklBQT/S04O3D33IV5pYa39y0aHCJudW49vO7eLnPAc8F+RoIiIA\nrPx8LWuPfka35M482Odu7VhxDXJSuvJgn7t5fuer/GH7n/m7Yd8hOS567zwoIlfn4MED5ObmceJE\nJYsXL6KgILi/kdJSbBGRJvac2MdbRe+SEp/M1wd+mfiYeKcjtXpDOl3H0apjvH/oQ/5U+ArfHvQo\nMe4Yp2OJSADuzJva7NHeYDhwYD833jiOjh3TyM8P/noQp3exEBEJG+XVFfxxxyu4cfG1AQ+TmtDB\n6UgRY0rPW7guoz97T+1n3r6FTscRkVamsrKCjIxMgJDsjKMCWUQEqGmo4ZntL1LTUMOXCu6iV/vu\nTkeKKG6Xm4f63EOX5GxWlazjk5J1TkcSkVZg5coVfPjh0r8UxydPnuDIkcNs2rQhqO2qQBYRAebu\nXUhZdblvD9/soU7HiUgJsW34+oCHSY5ry7y9CzlaVeZ0JBEJczExsZSVlTFr1j0ApKZ25Kc/fYJJ\nk6YEtV0VyCIS9QqP7+LTY5vISenCjNzgvulGu7TEjjzQZzYN3kZe2vUGjZ5GpyOJSBgbM2Ys9933\nIDk5PULargpkEYlqVfXVvLbnLWJdMTzY5x4tHguBAel9GZE1hMNnP2fZ4ZVOxxER+V9UIItIVJu7\ndyFn6s5ye89btUdvCM3Kn0b7+HYsPriMknNHr/wEEZEQUoEsIlFrW8UONpRtpnu7bkzIGet0nKiS\nFJfEfQV30eht5OXdb2qqhYiEFRXIIhKVztVV8fqet4l1x/JQn7s1tcIB/dP7MDJ7KEfOlrC0+COn\n44iI/IUKZBGJSm/unc/Z+nNM7XkrWW07OR0nat2VN40Obdqz+NByjpwtdTqOiAigAllEotCW8kI2\nlW+jZ7scTa1wWFJcIvcXzMLj9fDy7jdo8DQ4HUlERAWyiESXs3XnmGPfJs4dy4N97sbt0tug0/qm\nGUZ3Hk7JuaN8cGiF03FERFQgi0h0eWPvfM7VVzGt12Q6tc10Oo74zcybSmqbDiwpXsHhs587HUdE\nopwKZBGJGpvKtrGlfDu92vfg5m5jnI4jTSTGJvBAn9m+qRa73qReUy1ExEEqkEUkKpypO8sbe98h\nzh3Hg31ma2pFGCromM+YLiMprTrGkoPLnY4jIlFM3yFEJOJ5vV7m2Heoqq/mjtzbyEzKcDqSXMLM\n3Cl0TEhl6eGPKT5zxOk4IhKlVCCLSMTbVLaVbRU7yG3fk3FdRzkdRy4jITaBBwp8Uy1e2q2pFiLi\nDBXIIhLRTtee4Y2984l3x2nXilbCdMxjbJdRHKsqY/HBZU7HEZEopO8UIhKxvF4vr9u3qW6o4Y68\nKWQkpTkdSZrpjtzbSEvoyLLijzl4+rDTcUQkyqhAFpGI9UnxZxQe30V+h16M7XKD03EkAAmxbXiw\nz2y8eHl595vUN9Y7HUlEoogKZBGJSKdqT/PC5jeIj4nnAU2taJXyU3O5qetoyqrLWXRwqdNxRCSK\n6DuGiEQcr9fL63veoqq+hjvzbic9saPTkeQqTc+9jYzEND48vIoDpw85HUdEooQKZBGJOBvLtrKj\ncg8DOhnGdB7pdBy5Bm38vwEAeGX3PBq0q4WIhIAKZBGJKDUN53m7aBFx7li+PvQBXC6X05HkGuV1\n6MmNXW6grLqcj46sdjqOiESB2FA3aIx5EhgJeIDvW2s3Njk3Efg50AC8b619whgzDpgL7ABcwHZr\n7fdCnVtEWofFB5dxpu4st/e8hczkdCpqzjodSVrAtF63srl8G4sPLWdop0GkJnRwOpKIRLCQjiAb\nY8YCedbaUcCjwFMXXfIbYCYwBrjVGFPgP/6xtXa8tfZmFcciciml547x8edrSE/oyC05NzkdR1pQ\nUlwSd+ROoa6xjneK3nM6johEuFBPsZgAzAew1u4BOhhjkgGMMT2BSmttqbXWCyz2Xw++kWMRkUvy\ner28uXc+Hq+HWb2nExcT53QkaWEjs4fQo10Om8q3sefEPqfjiEgEC3WBnAVUNHl83H/si86VA9n+\nz/saY+YbY1b5p2GIiPyNTWVb2XfqAP3T+jAgva/TcSQI3C439/SegQsXc/cu0II9EQmakM9Bvsjl\nRoYvnNsL/MxaO9cY0wv4yBiTa6297DtjRkZKS2WUVkp9IHrU1J9n/rrFvoV5I79ERvJf/+3VDyJL\nRkYfJp4cw7L9n7Dh5EamF9zSjOeoD0Q79QEJVKgL5FL+OmIM0Bk42uRcdpNzXYBSa+1RfIv0sNYe\nMMYc858rvlxDFRVamBPNMjJS1AeiyNtFizhZc5opPSbirkn4y8I89YPIdEvnCaw9vIm5OxbRJ7kP\nHdq0v+S16gOiPiAQ+A9JoZ5isRSYBWCMGQyUWGurAKy1xUCKMSbHGBMLTAWWGmPuM8b8yP+cLCAT\nKAlxbhEJU0eryvjoyGrSEjpyS/ebnY4jIdA2Lok7cm+jVgv2RCRIQlogW2vXAZuMMWuAXwOPG2Me\nNsbc4b/km8AcYCXwurW2CFgIjDPGrALeAb5xpekVIhIdfAvzFvgW5uVPI14L86LGDdnD6N6uGxvL\ntrL35H6n44hIhHF5vV6nMwSDV79OiW76lVp02FS2jed3vkq/tAK+OfAr/+umIOoHka34zBF+tfG3\nZLXN5B+HfZ8Yd8z/ukZ9QNQHBCAjIyWgHdF0Jz0RaZXON9TydtEiYl0xzMqfrjvmRaHu7boxqvNw\njlaV8fHna5yOIyIRRAWyiLRKSw59yKna09zS/SYyk9KdjiMOmZ47mbaxSSw+uIzTtWecjiMiEUIF\nsoi0OseqyvnwyCo6JqRyqxbmRbXkuLZMy53M+cZaLdgTkRajAllEWhWv18vcv1mYF+90JHHY6M7D\nyUnpwoayLew7ecDpOCISAVQgi0irsqWikD0n99G3o2Fgej+n40gYcLvc3N17JgBv7p1Po6fR4UQi\n0tqpQBaRVqO2sY639r1LrCuG2b21ME/+qmf7HEZlD6O06hirStY5HUdEWjkVyCLSalxYmDchZxyZ\nSRlOx5EwMz33NpJiE1l0YCmna7Wtl4hcPRXIItIqlFWV8+HhVaS26cCkHuOdjiNhKCU+mWm9JnO+\n8Tzz92vBnohcvasqkI0x/+H/2N0YM7xlI4mI/C2v18vcfQtp9DYyK38abbQwTy5hTJcRdEvpwmfH\nNlN06qDTcUSklWp2gWyMSTPGtPM/fNcYUwB8FxgblGQiIn7bKnaw+8Re+nTszXUZ/Z2OI2HMt2Bv\nBqAFeyJy9WIDuHYSYIwx8YAXGAz8I1AYjGAiIgB1jXXM2/cuMa4YZve+Qwvz5Ip6te/OyOyhrD+6\nkaVFqxiaOtTpSCLSyjR7BNla+xrwc2vtP1prfwJ8A8gHfhSscCIiyw6v5GTtKSbkjKWTFuZJM83I\nnUJibCJv7niXc3VVTscRkVYm0DnILxtjRvs/7wwcstb+RwtnEhEB4OT5Uywr/ph28SlM6q6FedJ8\nKfHJTOk5kar6Gt47uMzpOCLSygRaIO8DfmSMmWatXQv8PAiZREQAWLB/CfWeeqbn3kZCbBun40gr\nM7bLDWSnZLK6dD2l5445HUdEWpFAC+QRwD3AQ8aYu4D5LR9JRAQOni5mQ9lmuqV0YUTWYKfjSCsU\n647loUGz8Hg9vF20CK/X63QkEWklAi2Q/9taWw/cC9wCfNbykUQk2nm9XubtexeAWfnTcbu0Zbtc\nncHZ/SlIzWf3ib3srNzjdBwRaSUC+q5jrV3s/9horf0G0O4KTxERCdjGsq0cOnOY6zMHktehp9Nx\npBVzuVzclT8NFy7eLlqkbd9EpFmuaVjGWvthSwUREQHftm7z9y8m1h3LjNwpTseRCNA5OYsxXUZS\nVl3BqpJ1TscRkVZAv7cUkbCy/PBKTtWeZny3G0lP7Oh0HIkQt/e8hcTYBBYfXMa5em37JiKXd8UC\n2RjT2f+xa/DjiEg0u7CtW0p8MpO63+x0HIkgKfHJ3NZjItUNNSw+uNzpOCIS5pozgrzQGNMG3x7I\nLmOMu+mfYAcUkeix8MAS6jz1TO91GwmxCU7HkQgzrusoMhLT+KRkHceqypyOIyJhrDkF7gGgChgH\nNAINTf7UBy+aiESTQ2cO89mxzXRL7szI7CFOx5EIFOuO5c68qXi8Ht4qWuR0HBEJY7FXusBaezeA\nMeY5a+1jwY8kItHG6/Uyb69vW7e78qdpWzcJmgHpfTGpeeyqtOys3EO/tAKnI4lIGLpigdzEN40x\nDwLDAC+wzlo7J9AGjTFPAiMBD/B9a+3GJucm4rs7XwPwvrX2iSbnEoAdwP+x1r4UaLsiEr42lW/j\n4JliBmUMID811+k4EsEubPv275/9mrf3LaIgNZ8Yd4zTsUQkzAQyTPMbYDpg8d1y+h5jzG8CacwY\nMxbIs9aOAh4FnvqCNmYCY4BbjTFNf7T/Z6AykPZEJPzVNdYzv2gxsa4YZuZpWzcJvi7J2YzuPJxj\n1eV8Urre6TgiEoYCKZD7W2tnW2t/Z639rbV2JhDo/V8n4L89tbV2D9DBGJMMYIzpCVRaa0uttV5g\nsf96/IVyAfBegO2JSJj78PAqTtae4uZuN5KemOZ0HIkSU3tNIiEmgcUHllFVX+10HBEJM4EUyPFN\nd60wxsQQ2BQNgCygosnj4/5jX3SuHMj2f/6fwA8BV4DtiUgYO1V7mqXFK0iJS2ZSj/FOx5EokhKf\nzG09J1DVUM372vZNRC4SSIH8HrDBGPOkfx7xRvyjwdfgcgWvC8A/73mttba4Gc8RkVZk4X7ftm7T\ncieRqG3dJMRu6jqajMQ0Vpas5VhVudNxRCSMNHsE2Fr7hDFmOTAC3yK9r1trPwuwvVL+OmIM0Bk4\n2uRcdpNzXfzHpgC9jDHTgK7AeWPMEWvtiss1lJGREmA0iTTqA+Ft/4liPj22ie4dujJ9wHjc7uDs\nXKF+IJfrA18eMptfrX6G9w4v4R/GPh7CVBJKeh+QQAU0RcJaux64lhUNS4GfAc8ZYwYDJdbaKv9r\nFxtjUowxOfgK46nAfdba3194sjHmp8DBKxXHABUVZ68hprR2GRkp6gNhzOv18txm3yY4M3pOobIy\nOLf+VT+QK/WB7nE96Z2ax+ajO1i5ZyN900wI00ko6H1AIPAfkkK62ai1dh2wyRizBvg18Lgx5mFj\nzB3+S77uo7GxAAAgAElEQVQJzAFWAq9ba4tCmU9EQmNz+XYOnD7EdRn96Z2a53QciWIul4tZ+dNw\n4eKtokU0ehqdjiQiYSDQRXbXzFr7k4sOFTY5txoYdZnn/muwcolIaNQ11jN//2JiXDHMzL3d6Tgi\ndEnOZlTn4awp/ZTVpZ8yruslvw2JSJRo9giyMWZyMIOISHRYceQTTpw/yc3dxpCRpG3dJDxM6zWJ\nhJg2vHdgKdXa9k0k6gUyxeK7xpgiY8y/GmO6By2RiESs07Vn+KB4BclxbZmsbd0kjKTEJzO5h2/b\nt8Xa9k0k6jW7QLbWTsF3m+li4GljzGJjzGz/fsgiIle0cP8S6hrrmNZrEomxiU7HEfkbN3UbQ7q2\nfRMRAlykZ609iW8R3WtAB+DHwDZjzMggZBORCFJ85gjrj238y3xPkXAT547lzrzb8Xg9vFX0rtNx\nRMRBgcxBHmuMeQHYhe8W01+11o7Atx3b00HKJyIRwOv1Mm/fQgBm5U/H7QrpBjoizTYwvR+9U/PY\nVWnZWbnH6Tgi4pBAvkv9AlgBGGvtD621u40xidbaQ8CbQUknIhFhU9lWDpwuZlDGAHqn5jodR+SS\n/mbbt33vats3kSgVSIF8zlr7srW2tsmxVQDW2n9v2VgiEilqG+t4Z/9iYt2xzMzTtm4S/rokZzOm\ny0jKqitYWbLW6Tgi4oAr7oNsjLkf+BeguzHmMODCd6vpeOBYcOOJSGu3vPhjTtWeZlL38aQndnQ6\njkizTO15KxvLtrL44HKGdxpMcnxbpyOJSAhdcQTZWvsq0Bff4rwx/j834tvRYnBQ04lIq3bi/EmW\nHV5J+/gUbu1+s9NxRJotOb4tt/e8hZqGGhYdXOp0HBEJsSsWyMaY31hrG4Fc4BXgZf+fV4GPg5pO\nRFq1+UWLqffUMz33NhJi2zgdRyQgY7vcQKekTFaXrKfk3FGn44hICDXnVtPP+z/+UzCDiEhk2X/q\nEJvKt9E9pRvDs/TLJml9Ytwx3JU/ld9ve555+97lu4Mew+VyOR1LRELgigWytXab/+PK4McRkUjg\n8XqYt28BALN6a1s3ab36pRXQL62AnZV72H58J9dl9Hc6koiEQHMW6X2Cb1HeF7LWjm3RRCLS6n16\ndBOHz5YwtNMgerXXnemldbsrbyq7T+zl7X2L6JtWQJy7Ob98FZHWrDn/yy83teKShbOIRKfzDedZ\neGAJ8e44ZuROcTqOyDXr1DaTcV1H8dGR1Xx05BMtOBWJAs0pkO+01n7vC0aSL2z3phFkEfmLD4o/\n4kzdWW7veQupCR2cjiPSIqb0mMiGY1tYcuhDRmQNpX2bFKcjiUgQaZGeiLSYiupKVhxeRWqbDkzM\nGed0HJEWkxSXxNRetzLHvsPCA+/zYJ+7nY4kIkHUnH2Qt/k/3Qj0A+4GZgMFwIbgRROR1uad/e/R\n4G1kZt4U4mPinY4j0qJGZQ+nc9ss3xz7M587HUdEgiiQpeXzgJFAIbAT381C3ghGKBFpfeyJIrZV\n7CC3fQ8GZ17ndByRFhfjjmFW/nS8eJm7byFer5bhiESqQJbitrPW3tbk8dPGmFUtHUhEWp9GTyPz\n9i3EhYtZ+dO1V6xELNMxj+sy+rOtYgebyrcxtNMgpyOJSBAEMoK8zxiTfeGBMSYL2NfykUSktVl7\n9DNKq44xMnsoOe26Oh1HJKjuzLudWFcM84sWU9dY53QcEQmCQPZBTgD2G2P2AB6gD7ApuPFEJNxV\n11fz7oEPSIhpw7Rek52OIxJ06YlpjM8Zy9Lij1h+eCVTet7idCQRaWHXug+y9nASiXKLDy2nqr6a\nGblTtPWVRI1J3W9m/dGNLC3+mBuyh2lLQ5EI05xdLFZe+ANU4BtN9gLxwP8Ncj4RCWPHqspZ+fla\n0hPTuKnbGKfjiIRMQmwC03tNpt5Tz/z9i52OIyItrNmL9IwxvwYmAVlAEZAL/GegDRpjnsS3G4YH\n+L61dmOTcxOBnwMNwPvW2ieMMYnAn4FOQBvgCWvte4G2KyIt7+2iRXi8Hu7Mu12335WoMyJ7CKtK\n1rKxbCvjuo6iV/seTkcSkRYSyCK9EdbaPsBWa+0w4BYgKZDGjDFjgTxr7SjgUeCpiy75DTATGAPc\nYowpAKYBG6y1NwH3AE8G0qaIBEfh8V3srNxD79Q8Bqb3czqOSMi5XW5m5d8BwJt2Ph6vx+FEItJS\nAimQa/0f2xhjXNbaTcDoANubAMwHsNbuAToYY5IBjDE9gUprbam11gu8D0yw1r5prb0wUp0DHAmw\nTRFpYXWNdby5dwFul5u7e9+hbd0kauV26MHwrMEcOVfKqpJ1TscRkRYSSIFsjTHfAlYBy4wxvyPw\nRXpZ+OYxX3Dcf+yLzpUDTbeVWwO8Anw/wDZFpIUtObSCE+dPMjFnHNltOzkdR8RRM/NuJzE2kXf3\nf8Dp2rNOxxGRFhBIgfwNYA7wE+B5fPOQp11j+5cbdvqbc9ba0cAdwKvX2KaIXINjVWUsP7yS1DYd\nmNxjgtNxRBzXLj6F6b0mc77xPG8Xvet0HBFpAYGsqkkC7gX64dvFohA4EWB7pfx1xBigM3C0ybns\nJue6AKXGmMFAubX2c2vtNmNMrDEm3Vp7/HINZWRou6lopz7Q8rxeL7/f8UcavY08OuxeumalOR3p\nitQPJBR9YGbaRDZWbGZj2Vam9BlH/04FQW9Tmk/vAxKoQArkefimQKzFN7p7IzCVwEaRlwI/A57z\nF74l1toqAGttsTEmxRiTg69Yngrc5//YHfiBMaYT0PZKxTFARYV+zRXNMjJS1AeCYMOxLews38uA\n9D70iO8V9n/H6gcSyj4wK/cOfrnxf/jDZ6/xk+E/IFY7u4QFvQ8IBP5DUiD/e9tZa29r8vhpY8yq\nQBqz1q4zxmzyzyduBB43xjwMnLLWLgC+iW8ahxd43VpbZIx5BviTv60E4FuBtCkiLaO6voa3it4l\nzh33l5X7IvJXOe26cmOXG1hVspblh1cxucd4pyOJyFUKpEDeZ4zJttYeBTDGZAH7Am3QWvuTiw4V\nNjm3Ghh10fXngfsDbUdEWtaigx9wtu4c03pNJj2xo9NxRMLStF6T2FKxnSWHljO00yD9XxFppa64\nSM8Y84l/9LYvsN8Ys9kYswnYD+QHO6CIOO/wmc9Z9fk6OiVlMjFnrNNxRMJWUlwid+ZNpd7TwLx9\nC5yOIyJXqTkjyP8U9BQiErY8Xg9z7Dt48XJP7xmaVylyBcM6Xc/a0s8oPL6bbRU7uS5DN9IRaW2u\nOIJsrV1prV0JrMZ3o467gDuBzv7jIhLB1pR+SvHZIwztNAjTMc/pOCJhz+Vyca+ZSYwrhrl7F1Db\nWOd0JBEJUCD7ID8FTAcsvrnHdxtjfhOUVCISFs7WnWPB/iUkxCRwZ961bnsuEj2y2nZiQs5YTtae\nYsmhD52OIyIBCuR3pf2tteOaPP6tMeaTlg4kIuHjnaL3qGmoYXbvO2jfRvuIigTith4T2Fi2leWH\nVzI8a7DuOinSigQyghxvjPnL9caYGAIrsEWkFdl38gCfHttEt+TOjO1yg9NxRFqd+Jh4ZudPx+P1\n8IZ9B6/X63QkEWmmQArc94ANxpgL845vxrdnsYhEmEZPI3P2voMLF/cW3InbFcjP0iJywcCMfgxI\n70Ph8d1sKNvC8KzBTkcSkWZo9nc9a+0TwONAMXAI+Lq19j+ClEtEHLTiyCccqypjdJcR9GiX43Qc\nkVZtdv4dxLnjeHvfIqrra5yOIyLN0OwRZGPMV6y1LwDrg5hHRBx24vxJFh9cRnJcW+7oNdnpOCKt\nXlpiR27rMYGFB5bw7oEl3GNmOh1JRK4gkN+b3mmMaR+0JCISFubte5c6Tz0z824nKS7J6TgiEWFC\nzlg6JWXyScl6is8ccTqOiFxBIAVyInDIGLPeGLPqwp9gBROR0NtxfDfbKnaQ274nI7KGOB1HJGLE\numO518zAi5c59h08Xo/TkUTkMgJZpPdvQUshIo6ra6zjzb0LcLvc3Gtm4nK5nI4kElF6p+YxrNP1\nbCjbwuqS9YztOsrpSCJyCVcskI0x7fDdbroA+AT4b2ttQ7CDiUhofVD8EZXnTzAxZxydk7OcjiMS\nkWbmTWVH5W4WHljCoMwBtIvX/uIi4ag5Uyx+7//4LNAH+Gnw4oiIE0rOHWV58cektunAbT0mOh1H\nJGK1b5PCtF6TqWk4z9y9C5yOIyKX0JwCuYe19u+ttYuAx4Abg5xJREKowdPAy7veoMHbyD1mBgmx\nbZyOJBLRbuwykp7turO5fDubyrY6HUdEvkBzCuT6C59YaxsB3QpIJIIsObSCI+dKuSF7GAPS+zod\nRyTiuV1uHup7N3HuON6w8zlde9bpSCJykeYUyBcXxCqQRSJE8ZkjfFC8gtQ2Hbgrf5rTcUSiRmZS\nBjPyplDVUM1re+bpNtQiYaY5u1iMMsYcbvI40//YBXittbrNlkgrVN9Yz0u73sDj9fBgn7tJjE1w\nOpJIVBnb5Qa2VexkR+Vu1h/bxA3ZQ52OJCJ+zSmQTdBTiEjIvXvwA45VlzOu62hMxzyn44hEHbfL\nzQMFs/nFZ08yb+9CTGouHRNSnY4lIjSjQLbWFociiIiETtGpg6w4/AmZienMyL3N6TgiUSstMZW7\n8qfz6p65vLp7Ho8P+ipuVyD38BKRYND/QpEoc76hlpd3vQHAg33vJj4m3uFEItHthuyh9E8rYM/J\nfawuWe90HBFBBbJI1FmwfzHH/TcE6dW+h9NxRKKey+XivoJZJMUm8k7Re5RXH3c6kkjUU4EsEkV2\nn9jLqpJ1ZLftxO29bnU6joj4tW/TjnvMTOo89by8+008Xo/TkUSiWnMW6bUoY8yTwEjAA3zfWrux\nybmJwM+BBuB9a+0T/uO/BMYAMcD/tda+E+rcIq1dTUMNr+ye69+D9R7i3CH/7y8ilzEk8zq2Vuxg\nS/l2Vhz5hIk545yOJBK1QjqCbIwZC+RZa0cBjwJPXXTJb4CZ+IrhW40xBcaYm4C+/ufcBvw6hJFF\nIsa8ve9yqvY0t/WYQE5KV6fjiMhFXC4X9/aeSUpcMu/uX0LpuWNORxKJWqGeYjEBmA9grd0DdDDG\nJAMYY3oCldbaUmutF1jsv34lMNv//FNAkjHGFeLcIq3a9oqdrD+2kZyULkzqPt7pOCJyCcnxbbmv\n4C4avI28vPsNGj2NTkcSiUqhLpCzgIomj4/7j33RuXIg21rrtdbW+I89Ciz2F9Ai0gzn6qp4zb5F\nrDuWB/vcQ4w7xulIInIZAzP6MSJrCIfPlvBB8Qqn44hEJacnIV5uJPhvzhlj7gC+AlxxZdFXn1hK\nY6Nq6GgWE+NSH/Cr7byBxnbniCvvx3+/dAA44HSkkFE/kNbaB7zuLFw9E3nvwHI+XFGPu7aD05Fa\nrdbaB+TShhVkcvf44N7gKtQFcil/HTEG6AwcbXIuu8m5Lv5jGGMmAf8ITLLWnm1OQzExmoUR7dQH\noD75cxrblRBT05E2p/NxReHfifqBtM4+EE9C2WBquq6hrvMmko7cjMur3/5crdbZB+RSEpPiychI\nCWobLq83dD9VGWNuAH5mrZ1kjBkM/NpaO7bJ+ULgdnyF8VrgPnxTLT4BJlhrm7s5pLeioll1tESo\njIwUor0PnK49w88/fZJ6Tz3/OPwHZCalOx0p5NQPpLX3gTfsO6wqWcctOTcxI2+K03FapdbeB6Rl\nZGSkBPRTUkjnIFtr1wGbjDFr8O1G8bgx5mH/9AmAbwJz8C3Me91aWwTcA6QBbxpjPjLGrDDGaAm+\nyGV4vV5e2/MWVQ3VzMy7PSqLY5FIcEfuFNIT01h+eCUHTh9yOo5I1AjpCHIIaQQ5ykX7iMG60g28\nsmcuBan5PD7oq7hd0XlPoGjvBxIZfaDo1EF+vfkZ0hM78o/Df0Ab3R4+IJHQB+TahfUIsogE37Gq\ncubtW0hCTAL395kVtcWxSKTI69CT8Tk3UlFTydy9C4jQgS2RsKLvnCIRpKbhPM8WvsT5xlq+VHAn\nHRNSnY4kIi1gWs9J5KR0Yd3RDawu/dTpOCIRTwWySITweD28vOsNyqrLGd/tRoZ2GuR0JBFpIXEx\ncTza/yGS49oyd+8CDpwudjqSSERTgSwSIZYWf8S24zvp3SGXGbla7S4SadISU3mk3/14vB7+WPgS\np2vPOB1JJGKpQBaJADsr97DowFJS23Tgkf736255IhHKdMxjRt4UTted5Y87XqHB0+B0JJGIpAJZ\npJWrqK7khZ2vE+OO4WsDHiIlPtnpSCISRBO6jWVI5nUcOH2It/YtcjqOSERSgSzSitU21vFs4YvU\nNNRwr7mTnHbaIlwk0rlcLu7vM5vObbNYVbKW9Uc3Oh1JJOKoQBZppbxeL6/unktp1THGdrmBG7KH\nOh1JREKkTUw8XxvwMImxibxu3+bwmc+djiQSUVQgi7RSHx5ZxabybfRq34O78qc5HUdEQiwjKY2v\n9PsSjZ5Gni18ibN155yOJBIxVCCLtEL2RBHzixbTPj6FR/s/QKw71ulIIuKAfmkF3N7zVk7WnuL5\nna/R6Gl0OpJIRFCBLNLKVNac5E87X8HtcvPogIdo36ad05FExEGTetzMwPR+7D1ZxIL97zsdRyQi\nqEAWaUXqGut5bsdLVNVXM7v3dHq17+50JBFxmNvl5qG+99ApKYMPj6xiY9lWpyOJtHoqkEVaCa/X\nyxz7NkfOlnBD9jDGdB7pdCQRCROJsQl8bcBDtImJ55Xdcyk5d9TpSCKtmgpkkVZiZclaPj22ie4p\n3bin9wxcLpfTkUQkjGS17cRDfe+l3lPPs9tfpKq+2ulIIq2WCmSRVqDo1EHe2vcuyXFteWzAg8TF\nxDkdSUTC0KCM/kzuPp7j50/w552v4/F6nI4k0iqpQBYJc6dqT/PHHS8D8Gj/B0hN6OBwIhEJZ7f3\nupW+HQ27TljeO7DU6TgirZIKZJEwdr7h/F/2N52Zdzv5qblORxKRMOd2uflKvy+RntCRJcUr2Hhs\ni9ORRFodFcgiYaq2sY6nt79A8ZkjjMgaws1dxzgdSURaiaS4JL428GESYhJ4cfcbbC0vdDqSSKui\nAlkkDNU11vOH7X+m6NRBrs8YwP0Fs7QoT0QC0iU5m8cHPUKsO5bnd75G4fFdTkcSaTVUIIuEmXpP\nA88VvoQ9WcSA9L58pd99xLhjnI4lIq1Qr/Y9+NbAR3C73Pyx8GV2VVqnI4m0CiqQRcJIg6eBP+14\nhV0nLH3TDF/t/4CKYxG5JvmpvfjGwC/jcrl4tvBF7IkipyOJhD0VyCJhotHTyAs7X6fw+C4KUvP5\nWv+HiHPHOh1LRCJAQcd8HhvwMF6vl2e2v0DRqYNORxIJayqQRcKAx+vhpd1vsLWikPwOvfj6wIe1\n17GItKh+aYZHBzxIg7eR32/7EwdOFzsdSSRshbxANsY8aYxZa4xZbYwZetG5icaYT40xa4wx/9Tk\neH9jTJEx5luhzisSbB6vh1d3z2Nj2VZ6te/ONwZ+hfiYeKdjiUgEGpDel0f63U+9p4Hfbf0TxWeO\nOB1JJCyFtEA2xowF8qy1o4BHgacuuuQ3wExgDHCrMabAGJPkv255KLOKhILX62WOfYf1xzbSPaUb\n37ruERJi2zgdS0Qi2PWZA3i4773UNtby261/5MjZUqcjiYSdUI8gTwDmA1hr9wAdjDHJAMaYnkCl\ntbbUWusFFvuvPw/cBhwNcVaRoPJ6vczdt4A1pZ/SLbkz3x70VRJjE52OJSJRYGinQTzY525qGs7z\n263PUXrumNORRMJKqAvkLKCiyePj/mNfdK4cyLbWeqy1tSHKJxISXq+Xt4sWsfLztXRum8W3r3+M\npLgkp2OJSBQZkT2ELxXcybn6Kp7a+ixlVeVORxIJG04v0rvcnQ90VwSJSF6vl4UHlrDiyCdkJWXy\n3eu/RnJcW6djiUgUGt15BPf0nsHZunP8ZsuzlFcfdzqSSFgI9R5Spfx1xBigM3+dOlEKZDc518V/\n7KpkZKRc7VMlQoRrH5i7YxFLiz8iOzmTn43/IamJ7Z2OFNHCtR9I6KgPXN5dGZNISIrlxa3z+O32\n5/jX8T8is22a07FalPqABCrUBfJS4GfAc8aYwUCJtbYKwFpbbIxJMcbk4CuMpwL3XfT8Zo8qV1Sc\nbZnE0iplZKSEXR/wer0sObSCRQc/ID2hI48PfJSGc24qzoVXzkgSjv1AQkt9oHmGdxzOqdwqFux/\nn58u/y++e/3XSE+MjCJZfUAg8B+SXF6vN0hRvpgx5hfAOKAReBwYDJyy1i4wxowBfgl4gXnW2v/2\nF9L/BXQH6oES4E5r7anLNOPVf4boFm5viLWNdby6ey6byreR2qYDPxj8TdISU52OFfHCrR9I6KkP\nBOb9g8tZdHApSbGJPNLvfvqk9XY60jVTHxCAjIyUgKbuhrxADhEVyFEunN4Qj9dU8mzhS5ScO0qv\n9t15tP+DtG/TzulYUSGc+oE4Q30gcGtLN/CGfZtGr4fpvSZzS/ebcLla77Ig9QGBwAtk3cdWJIh2\nVVpe2Pka1Q01jO1yA3flTyNWt48WkTA2qvMwOid34rnCl1lw4H0On/2cB/rMJiE2weloIiHj9C4W\nIhHJ6/XywaEV/H7b89R56nmgYDb3mJkqjkWkVejRLod/GPY98jv0YktFIb/a9DvKqyuu/ESRCKEC\nWaSFnW84zx93vMLCA0to36YdPxz8TW7oPMzpWCIiAUmJT+Y7gx7j5q5jOFZVxi83/g+Fx3c5HUsk\nJFQgi7SgsuoKfrXpd2ytKCS/Qy/+Ydj36N6um9OxRESuSow7hlm9p/Nw33tp8DTwzPY/897BZXi8\nHqejiQSVft8r0kIKj+/izzvncL7xPDd3G8PM3NuJccc4HUtE5JoNzxpMdttOPFv4EosPLuPI2c95\nuO+9JMYmOh1NJCg0gixyjTxeD+8dWMoz2/9Mo7eRh/vey6z86SqORSSidEvpwv839LsUpOZTeHw3\nv9z4PxytKnM6lkhQqEAWuQY1DTX8YfuLLD60nLSEVH405HGGZw12OpaISFAkx7flW9c9wi05N1Fe\nfZxfbfwftpYXOh1LpMWpQBa5SqXnjvHLDf/DjsrdFKTm8/fDvku3lM5OxxIRCaoYdwwz8qbwSL/7\n8Xq9PLfjZRbsf59GT6PT0URajOYgiwSopqGG9w9+yEefr8bj9XBLzk1Mz52M26WfN0UkegzpdB1Z\nbTN5tvAllhZ/ROHxXczKn05Bx3yno4lcM91JTyJSMO6c5PF6WH90Iwv3L+Fs/TnSEjoyu/d0BqT3\nbdF2pOXoDlqiPhB81fXVzN+/mLWlG/Di5br0ftyZP5X0xDSnowHqA+KjO+mJBMGB04eYu3cBh8+W\nEO+OY1qvSUzoNpa4mDino4mIOCopLon7CmYxpstI5u5dyLbjO9l5wjKh21hu7X4zCbFtnI4oEjCN\nIEtEaqkRg1O1p5lftJgNZVsAGNppEDNyp5Ca0OGaX1uCTyNHoj4QWl6vl01lW3ln/2JO1Z6mfXw7\nZuRNYVin63G5AhrAazHqAwIaQRZpEfWN9Xx4ZBUfHFpBnaeenJQuzMq/g9wOPZyOJiIStlwuF0Oz\nrmdARj+WFX/EssMreXHXHFZ9vo7ZvafrxknSamgEWSLS1Y4YeL1eth3fydv7FlF5/gQpcclMz53M\nyOyhWoTXCmnkSNQHnFVZc4K3i95ja0UhLlyMzB7K9NzJtItPCVkG9QEBjSCLXLXSc8eYt28h9mQR\nbpeb8d1uZErPibpTlIjIVUpL7MhjAx5k78ki5u5dyLqjG9hSvp3bek7kpq6jiXWrDJHwpBFkiUjN\nHTHwer0cOnOE1SXr+axsMx6vh75phll50+jUNjMESSWYNHIk6gPho9HTyJrST1l0YClVDdVkJqYz\nrttohncaTFJc8AYi1AcEAh9BVoEsEelKb4jV9TVsKNvCmtJPKTl3FIDMpHTuyptG//Q+oYopQaZv\njKI+EH6q6qtZdGApq0vX4/F6iHPHMSTzOkZ3GUHPdjktvphPfUBABfIFKpCj3Be9IXq9Xg6eOcya\nkk/ZVL6Nek89bpeb69L7MbrLCExqnuYZRxh9YxT1gfB1pu4s649uZE3pZxyvqQSgc9ssRncewfCs\n60mKS2qRdtQHBFQgX6ACOco1fUOsrq/ms2O+0eLSqmMApCd0ZHSXEYzMHhrSxSISWvrGKOoD4c/j\n9bD35H7WlH7KtoqdNHobiXPHMjjzOkZ3HkGv9t2vaVRZfUBAi/REAN9o8f5Th1hT+imby7dR72kg\nxhXD4MyBjP5/7N13fFRV+vjxz0x6IwlJgNAhhCf0rihNERdFBLuiotjrqqvud1d/X/vuurtfe2VX\nBdEV6yoKoiKKVEUIhp4DgSR0SCOVtJn5/XEnOCAlCZOZlOf9esWZueXcZ5jjnWfOPfec9qfTMzZJ\nW4uVUqoRsNvspLROJqV1MsWVJe5W5ZWs3JfKyn2ptItoy8j2p3Nau8FEeKlVWamT0RZk1WxUO6vJ\nKtrJloIM1uVtYGeR1bc4ISyOEe2t1uKo4Eg/R6l8SVuOlNaBpsnpcrK1YDvL96wkLWcDDpeDQHsg\nAxP60ru10DM2qdYTNmkdUKAtyKoFcTgd7CjehSnYxtaCbWwrzKLKWQVAgD3Auumj/ekkx3bX1mKl\nlGpC7DY70roH0roHxZUlrNyXyvLdK1m9P43V+9MAiA+Lo2dMEj1jrb/okFZ+jlo1J9qCrJoMp8vJ\nzuLdbCnYxpaD29h2MJMKR+Xh9YkRba0TZUwSZ/QYwKEipx+jVY2BthwprQPNh8vlYlfJXrYWZLDl\n4Da2FmRS7ig/vL5teJvDyXJyTPfDVwy1DihoAjfpichzwHDACdxnjFntsW4c8FegGvjKGPOXk+1z\nHJogN3Eul4uiyhIOlB2wkuKD28g4mMmhas+TYQLJ7oS4Z2zSEd0n9ISoQOuB0jrQnDmcDnaV7LEa\nTYYvEHsAACAASURBVAq2kVGYSaVHo0n7iHYkxyYxuHMvwqtbkRAWR1BAkB8jVv7UqLtYiMhooIcx\n5kwRSQFmAGd6bPIicC6wF1gsIp8AbU6yj2rCDlWXk1OWy/6yHA6U5XDgUK71WJZLuaPiiG3jw+IY\nlNDfah2I7U5MSLSfolZKKeVvAfYAurTqRJdWnTi3y1k4nA6yi3exxaPb3Z7SfSzetRwAGzZiQ2No\nG55Am/B42oS5H8MTaB0ao13x1BF83Qf5HGAOgDEmXURiRCTSGFMiIt2APGPMHgAR+RIYByQcbx8f\nx67qwOVyUemsoqSylNKqUkrcf4UVRRwoy+XAISsJLqr8bctOoD2QNmHxh09c7cLbkBzbndahsX54\nJ0oppZqCAHsA3aO70D26C+d1HUuVs5rsop3kOPaTmbPL+u4py2Fz/hY25285Yt9AeyAJYXG0CU+g\nTVg88WGtiQyOJDIogsigcCKCIogICtckugXxdYLcDvDsHpHrXpbhfszxWJcDJAFxJ9hHeZnT5aTS\nUUWV0/3nqKLS/bxmec1jhaPSnfyWWY+VvybCpVWlVDmrj3scGzZah8bSq3VP64QUHm/9qg9LIDY0\nWk9CSimlTkmQPZAeMd04I6E/OXG/NsaUV5e7r1b+esXygPtK5t7S/cctz4aN8MAwIoLDiQyKICIo\nwp1ARxAZHEFYYChB9iCCA4KtR3vg4edB9iCCAgIJtgcTFBBEoC3A6zMGKu/y9ygWJ6odx1t30hr1\nzPJ/UVlx/OTMX37T2/sY/b+PXuLCZf3X5fHK5XI/s57XrPFc7nQ5cbqcOFxOXO5Hp8uJEydOpxMn\nLvd6By6XC4fLQZWjimqX45TeY0hAMJFBEbSPSDx8Evn1RBJOVHAUbcLjiQ9trX3BlFJK+VxoYCid\nozrSOarjEctdLhfFVSUcKMslv7zAauw53PBTdsTV0NxD+Thd9b8R3IaNoIAggmyB2Gw2Amx27LYA\n7DYb9qOeB9js2GoesWO32bBhA5sNm7ssANvh5dYyW82RbL++OjKGoxfYTry+kQiwBXBul7PpFNW+\nQY/j6wR5D1brb432WP2Na9YleqzrAOwGKk6wzzE9OOK2xvq5Kh9KSNAZ8pTWA6V1QNW+DrShFUk0\nbOKlmgZfX8deAFwGICKDgd3GmFIAY0w2ECUinUUkEJjo3v7b4+2jlFJKKaWUt/ljmLe/AWMAB3AX\nMBg4aIz5XERGAv/E6mnwiTHm+WPtY4xZ79OglVJKKaVUi9FcJwpRSimllFKqXnSoAKWUUkoppTxo\ngqyUUkoppZQHTZCVUkoppZTy4O9xkL1KRJ4DhgNO4D5jzOqT7KKaCRHpizXj4nPGmNdEpCPwLtaP\nwL3AVGNMlT9jVA1LRP4JjAQCgL8Dq9A60KKISBjwNtAWCAH+AqxF60GLIyKhwAbgSeB7tA60GCIy\nBvgY6/O3AeuA/6OOdaDZtCCLyGighzHmTOBm4CU/h6R8RETCsT7vhR6LnwReNsaMAbYBN/ojNuUb\nInIW0Nv9///5wAtYdeAVrQMtyoXAKmPMWcCVwHNoPWipHgHy3M/1+6Dl+cEYM9YYc7Yx5l7qUQea\nTYIMnIPVgogxJh2IEZFI/4akfKQcKynynEDmLGCu+/lcYJyPY1K+tRi43P38IBCBNTTkF+5lWgda\nAGPMR8aYZ9wvOwM70XrQ4oiIACnAl1gtiGPQ74OW5ugJ486ijnWgOXWxaAd4dqnIdS/L8E84yleM\nMU6gwjonHhbhcfnkAEfO0qiaGWOMCzjkfnkT1hfjeK0DLZOILMeajfVC4FutBy3Os1jzLExzv9bv\ng5ant4jMAVpjtR6H17UONKcW5KPpdNOqhtaFFkJEJmNdOrubIz93rQMtiDFmBDAJeA+tBy2KiEwF\nVrhn5z0WrQPN31bgcWPMRVg/kt7iyAbhWtWB5pQg78FqMa7RniMvuauWpVhEQtzPO2DVD9WMich4\n4CHgPGNMMVoHWhwRGey+QRdjzDqsGza1HrQsFwCTReRHrKtJjwAlWgdaDmPMHmPMx+7n24F9QGxd\n60BzSpAXAJeBdZIEdhtjSv0bkvKjhcCl7ueXAl/7MRbVwESkFdY09RONMYXuxVoHWp7RwAMAItIW\niMSqB5e512s9aOaMMVcZY043xpwBvIl1eV3rQAsiIleLSM15oB3WqDYzqWMdaFZTTYvI37A64zuA\nu4wx6/0ckvIB9w+iZ4EuQBWwG7gGmIU11FM2cIMxxuG3IFWDEpFbgMeALViXz1zA9ViX1rQOtBDu\nob3eAjoBocDjQCrW8E5aD1oYEXkMyAS+QetAi+EeoGE2EAMEYZ0H1gLvUIc60KwSZKWUUkoppU5V\nc+pioZRSSiml1CnTBFkppZRSSikPmiArpZRSSinlQRNkpZRSSimlPGiCrJRSSimllAdNkJVSSiml\nlPKgCbJSSimllFIeNEFWSimllFLKgybISimllFJKedAEWSmlGgkRGSYi20Sk6zHWLRaRC49aFioi\n+SLS4QRljhGRpQ0QrlJKNVuaICulVCNhjFkF7DbGZB1j9VvAtKOWXQz8aIzZfZKiXacenVJKtRyB\n/g5AKaWUxd1yvO04qz8G/k9EYo0xBe5l1wFvuPe1AdMBAUKAlcaY+zzKHgP8xRgzyv16JrDUGDND\nRO4GLsf6TkgH7gRaA++5dw8D/mWMedtLb1UppRo1bUFWSqnGYzSwSERGiMiLIvLHmhXGmEPAp8AU\nABFJBAYAX7g3iQXWGmPOMsacAYwXkd5Hlf+blmQRGQZcbIwZY4wZARQCtwBXApuNMWOBMUC4N9+o\nUko1ZtqCrJRSjcdoIBP4Erj/GOtnAK8CrwHXALONMdXudQeBziKyAqgA2gHxtTjmWUCSiHwP2LAS\n4Ur3seaKyAxgPvDver4npZRqcjRBVkqpxqMbMBd40Rhz7dErjTGrRCRERFKAqcBVHquvAoYCI4wx\nLhFZddTuR7ceB7sfK4AvjDH3HH08dwv0GOAK4D5gZD3ek1JKNTnaxUIppRoBEWkH7DfGfA50FZEg\nETn/GJu+BTwClBpjNnssbwsYd3I8BEjC6otcowjo6D5WOHC6e/ly4HwRiXCvu0NEhovIFOA0Y8z3\nWH2SO4mIfmcopVoEPdkppVTjMBj4zv18MVaL8KJjbPcecAnw5lHLPwbOFJFFWKNbPAO8BMQAGGPS\ngHUikgrMxEqMMcakYnXb+EFElmC1GKcBm4Dn3OV9D/zdGOP0zltVSqnGzeZy6eg/SimllFJK1dAW\nZKWUUkoppTxogqyUUkoppZQHTZCVUkoppZTyoAmyUkoppZRSHjRBVkoppZRSyoMmyEoppZRSSnnQ\nBFkppZRSSikPmiArpZRSSinlQRNkpZRSSimlPGiCrJRSSimllAdNkJVSjYKIOEXko2Msf1NEnO7n\nw0Tkq5OU00ZELmyoOOtDRCaJyB4RefU4668RkVQR2SQiRkTeE5GuJyhvlohccJJj3iUiT5xCzNeL\nyLfHWN7F/VnVxJotIu+LSEod47tCRCKPs+5vInKr+7lTRNrXMfaeIjLK/fwiEXmzLvsrpVSgvwNQ\nSikP/UUk0hhTAiAiQcBQwAVgjFkFnH+SMsYC5wBzGzLQOpoEvGGMeezoFe5E8AFgsjEm3b3s98By\nERlojMk5eh9jzPUnO6Ax5pjJeB25jrO82hjTG0BEbMBtwBIRGWGM2Vqb+IAngOVAydErjDEP1yKG\nE7kY6/ttqTFmDjCnHmUopVowTZCVUo3JIuAS4B336/HAKqAfgIicBbwBCLAaeNIYM0dEugMrgIuA\nl4EAEYkA/gW8aYxJdu8/pua1iDwGdAAGAO8ZY14SkUeBq4EQrKTqfmOMS0QuBx4FAoBK4B5jzBLP\nwN2J4l/c8buAn4C7gZuBy4AKEWlnjLntqH2eAK6uSY4BjDEvi8ho4F7gf0VkEVYyeTFwE/A0VsI9\nW0SmuV/vA14AZhpj7DXvzxhzq3v/L9yxdQOWGGOudscwyR13MFAM3GSMWVfLzwtjjAuYLiIdgMeA\na93Hq4nvL+73bwN2AlPdxxNgkTv+W4B8rB82TwETga3GmL+597taRK4DWgH/MMa8LiLXA9caY851\nv4/rgWuBF4GH3P/eMcCGmu1EJBaYjvWZVwPvGGP+6d7fCVwH3A+0Bf7PGPOCux69C6S4/42+A+40\nxjhq+2+klGp6tIuFUqox+QgrQa0xxb2shgtwGWOcWEnVP0QkBHgGeNQY8xPwCvBJTQLIb1sgPV+f\nD5znTo6nYiVyQ4Ek998d7u1eBc53t5reidUifLQrsRL6QUAfIBa4zxjzEvAZ8KJncuyWArQyxiw6\nRnlzgTEerwcbY/q43yMA7oTvVWCsMWaQ+/jHa3GdiJWA9gTGisgZIhIAzMRKilOwkuhnjrP/yXwB\nnO25QER6A5cDvY0xgvXvcI4x5ib3JmOMMSvcz8cCw4wx/z1G2Z2NMf3d7+9ZEYlzL//NZ2uMmcev\n/95/PGq7p4F893sdBdwpImd67N/bGDMYmAz8zf0D5nqgwP3Z98RKrPvU5h9EKdV0aYKslGosXMAP\nQB8RiReRMOAM4HusVsQjGGNSgXnAx0CCMebf9TjmSmNMgfv5RGCGMabEnYC/hdXiCrAfuENEOhtj\nVhhjHjxGWROAWcaYcner6kzgdyc5fmsg9zjr9rvX15h/jG1OB4wxZrP79esnONYnxphKY0wZsAUr\n6XQAbdxdVwCWAd1PEvPxFAHRRy07CMQDU0UkxhjzqjHmPx7rPT/X74wxVccpexZYbxTYDAypZ4wT\ngNfcZRUAn3LkZ/Su+3EN1lWENsAB4AwRORcINMbcVZcWdqVU06QJslKq0XAnlp9itcZOBL45yaXs\n193bvVXPQ+Z7PI8BHnTffLYZ+D8g1L1uEpAIpLpvpht9jLISgAKP1wVYCdaJ5GJdzj+WtljJ2bFi\nrRF71PLdJzhWocdzB1Z3EYD7RGSt+z3PpP7fC105Ml6MMXuwfmRcDuwQkbnurhjHcqz3V8PzR0QR\n1vuuj5N9RoUA7h9INiDAGPMJ8DxW14/9IvKyu2+8UqoZ0wRZKdXYfIDV1eEy93M4freBp7GSl4fd\nLc5H80wE4cgW2aPtAf5mjOltjOlljOlpjBkJYIzJNMbcaIxJAF4CZh9j//1AnMfrOPeyE9mClThO\nPMa6C4GFJ9m/CIjyeF3X0R7OAP4HmGiM6YXVX7q+LgO+OXqhMWaxMWYiVsK/E/h7Pcr2/NxqfhQc\n/dnWJmmuz2eEMeYNY8xwoDdWF5zranEspVQTpgmyUqqxsAEYY37Eaq3ta4xZ7LnOk3sYsfbGmAeA\nr7Fa+ACqsFqDAfYCie4uGwEc2b/5aJ9jdQUIc5d/q4hMde+7QERqEtGVgPMY+8/DukEtTEQCsW6m\nm3eiN+xuMX8YeFlEBnq8t7uBwVj9i08kFegnIt3d/WVvOsn2R2uDlSDuEpFwrP62EbXY7/DnISJ2\nEbkTqyX/r54bici5IvKKiNiMMYeAtfz6Y6eaXz+nk6m5oTAFq2/4KqzPVkQk2B37ZR7be9YBT/OA\nmuHj4rFat0/4GYnI/4rIDQDGmL1AJvUbWUMp1YRogqyUaiw8k45PObL19IiExJ0QvQjc5V70KDDF\nnWQuAM4RkZXGmG1Y3QbSgCWcoEXWPRzYXGCNiGzCasH9xhiTC3wFrBKRDVitxzceY/9PsPoJpwLr\ngB1YI2r8Jv5j7Pc/wFsisllEDFbf6zHGmIPH2b9m2Lt9WAn2D1ijZizh2I53o+LXWInmNvfz54FC\nEfn4RDEDdo+uKLuAc4HRxphdR5W/GAgHtojIeuAKrM8KrJsvV4jIZSc5lgvIEpFfsD6H37v/XRZh\n/VjZAnzJkUO5zQVuF2tcbc+y/xdo7Y77B6wrBqlHxcxRr9/F+uG02V0vKvi1r7JSqpmyuVy+/SEs\nIs8Bw7FaYO4zxqz2WDcOqwWiGvjKGPMX9/JrgD9itQo8aow54UQBSinVErlHjVhqjIk76cZKKaWO\ny6fjILtvbOlhjDnTfalsBuA5xM6LWC0Re4HFIvIJ1k0fj2INnRSFNWaoJshKqRbP3W1kB3CxMeZn\n4CrgR/9GpZRSTZ+vJwo5B/dlMGNMuojE1MyaJSLdgDz3Xc+IyHz39jnAt+6hicqA230cs1JKNUrG\nGIe7/+8sdx/kvdS9H7JSSqmj+DpBboc1+1WNXPeyDPej55SqB7BuxogAIkTkc6ybLp4wxnzvm3CV\nUqpxM8Z8jnWDoVJKKS/x91TTv7kz/ah1Lvdja6wpZLth3ZjRpeFDU0oppZRSLZGvE+Q9WC3FNdpj\nXRKsWZfosa6De1kpsMI9HNJ2ESkWkXj3neXH5HK5XDbbiXJvpZRSSinVgtQpMfR1grwAeBx4Q0QG\nA7uNMaUAxphsEYkSkc5YifFErLEvy4CZIvJPrJbkiBMlxwA2m42cnOIGfBuqsUtIiNI6oLQeKK0D\nSuuAAqx6UBc+TZCNMT+6p2ldjjUL0l0icj1w0N2P7g6smbNcwPvGmAwA92gWP7mX3+3LmJVSSiml\nVMvi83GQfcSlvxZbNm0xUKD1QGkdUFoHlCUhIapOXSx0Jj2llFJKKaU8aIKslFJKKaWUB02QlVJK\nKaWU8qAJslJKKaWUUh40QVZKKaWUUsqDJshKKaWUUkp50ARZKaWUUkopD76eSU8ppZRSSrVgL7/8\nHBs3bsBms3HvvQ+QktL7iPWvvfYi69atxeFwcO2102jVqhWPPPJnundPwuVykZSUzH33PdigMWqC\nrJRSSimlfCItbQ27du1k+vQZZGdn8fTTTzJ9+ozD69esWU1WVibTp8+gqKiQG264hkceeZJBg4bw\n1FN/91mcmiArpZRSSqlTNm/e53zzzXxsNhsulwubzca0aTczePDQw9ukpq5i1KizAOjSpSslJcWU\nlZURHh4OwKBBQ+jTpy8AkZFRVFSU43Q68fXMz5ogK6VUI+V0udiUlc+StXvZuusg1PH7ITIsiOF9\n2nJm30Rio0IaJkilVIP76PsMVqUf8GqZw1LacMXYHifcZsOGdSxdupikpGRCQoIpLCxk0qSLj7v9\nxImTmThx8gnLzMvLRaTX4dfR0THk5+cdTpBtNhshIaEAzJ07h+HDR2Cz2cjKyuShhx6gqKiIadNu\nZtiw02v7VutFE2SllGpk8ovKWbZuL0vX7SWvqByA2KgQgoMD6lROzsFD/Hfxdj5bkkn/pDhGDUik\nf1IcAXa9P1spdXIulwuHw0GXLl0RSeGee25n4sTJvP76y9x1171eO8axLF36A/Pnz+X551+hrKyM\nG2+8lbFjx7F79y7uued2PvxwDoGBDZfGaoKslFKNQLXDSdrWXJas28PG7fm4gJCgAEYPSGTUgPZ0\nT2yFzWarU5ll5dWs3LyfJWv3kJaRS1pGLtGRwYzsl8jI/om0jQ1vmDejlPKqK8b2OGlrb0Po128A\nb7/9FiIpFBUVUV1djd1uJyws7Jjb16aLRXx8Avn5eYdf5+bmEBcXf0Q5K1f+yLvvvs1zz71CeHgE\n4eERjB07DoAOHTrSunUcubk5tGuX2ADv2qIJslJK+dGe3FKWrtvDig37KC6rAiCpQytG9W/PsJQ2\nhIXU/zQdHhrI2YM6cPagDuzYX8zStXv5ceM+vvwxmy9/zCalcwyjBrRnSM8EgoPq1jqtlGr+qqqq\nqPldvnz5EsaPn0BpaQlVVVWkp29i//59jBkz9vD2telicdppw5kx499MmnQxxqSTkNDmiIS7tLSE\n1157iRdffJ3IyEgAFiz4mry8XKZMuZa8vFwOHiwgPj7B+2/YgybISinlYxWVDn5O38/StXvJ2F0I\nWP2FfzesE6MGtKdDfITXj9m5bRTX/C6Ky89OInVLDkvX7iF9x0HSdxzkvZBAhvdpy+gB7encNsrr\nx1ZKNU3p6ZsICgpi2bIl5ObmMnXqNJYtW0xUVBQVFZWMHDmmzmX27dsfkV7ccceN2O0B3H//n8jP\nz2PGjH/z4IMP8d1331JUVMijj/75cCv0//t/j/Pcc/9g2bLFVFdX8+CDDzVo9woAm6/vCvQRV05O\nsb9jUH6UkBCF1gHV2OqBy+ViwaqdfL4sk/JKBzagd7fWjB7QnoE94gkK9G3f4P0FZSxbt5dl6/dS\nWFIJQI+O0dx0Qa9m0/2isdUB5XtaB+pv9ux3SUnpdUQXidmz32XKlGuZNestxo49l86du/gxwtpL\nSIiqUx81TZBVs6QnRAWNqx6UHKpixpebScvIJSo8iLMHdWBk/0Tio4/dl8+XHE4n67fl80PabtZt\nyyM0OIAbJvRiWEobf4d2yhpTHVD+oXWgfnbv3sUjj/yJK664mvPOu8Df4ZwyTZAtmiC3cHpCVNB4\n6sG2PYVMn7ORvKJyenWJ5dZJfYiOCPZ3WMe0YsNe3vnGUFnlZOzgDlw5NtnnLdve1FjqgPIfrQMK\n6p4gax9kpZRqIC6Xi29X7eTjH7bhdLq4aGQ3Jp7ZFbu9bqNR+NKZfRPp2q4Vr8/ZwPdrdrNtdxF3\nXNSHNs2ky4VSStVG020WUEqpRqy0vIpXPl3PB99nEBEWxINXDWTSyG6NOjmu0T4+gv+9figj+yeS\nvb+YJ95exWovT1KglFKNmbYgK6WUl23fU8TrczaQV1ROSucYbpvUh+jIpjWTXUhQADdO6IV0iuHd\nBYbX5mxg3JCOXH52jybd5UIppWpDE2SllPISl8vFwtW7+GhRBk6ni0kjujJpRNNoNT6eEf0S6Zpo\ndblYmLqLjN2F3HFRXxJi/H9zoVJKNRRtBlBKKS+o6VLx/ndbiQgN5P6rBnLRqO5NOjmu0SE+gkeu\nG8qIfu3I2lfM4zNXkWpy/B2WUko1GG1BVkqpU5S51+pSkVtodam4dVIfYppYl4qTCQkO4KYLeiOd\nYvnPAsOrn61n3NCOXHF2DwIDtK1FKdW8aIKslFL15HK5+C51Fx9+b3WpuPDMrkxuIjfi1dfI/ol0\nS4zitTkbWLh6F9t2F3LH5L7Ea5cLpVQzoj/7lVKqHlwuF+8v3MrshVsJDw3k/isHcvHo5tGl4mQ6\nJETy6PXDOLNvOzL3FvPkrNXszi31d1hKKeU12oKslFL1MGdpJgtTd9EhPoL7rxxIbFTz6lJxMiHB\nAdw8sTfdElvx3rdbePaDX3jo2iF6855S6qRefvk5Nm7cgM1m4957HyAlpffhdb/8ksojj/yZ7t2T\ncLlcJCUlc999D/LSS8+yadPGY+7TEDRBVkqpOvp65Q7mrsiiTUwYD1w1sNn1N66Lc4Z0pNrh5MPv\nM3jmg1/48zVDWtyPBaVU7aWlrWHXrp1Mnz6D7Owsnn76SaZPn3HENoMGDeGpp/5+xD67d+864T7e\n5vMEWUSeA4YDTuA+Y8xqj3XjgL8C1cBXxpi/iMgY4GNgA2AD1hlj7vV13EopBbA4bTcfLcogNiqE\nB1t4clxj/GmdOVRRzRfLs3juwzT+dM1gIsOC/B2WUsrH5s37nG++mY/NZsPlcmGz2Zg27WYGDx56\neJvU1FWMGnUWAF26dKWkpJiysjLCw3+drdPlch1Rbm328TafJsgiMhroYYw5U0RSgBnAmR6bvAic\nC+wFFovIJ+7lPxhjrvBlrEopdbSVm/bzzteGSPfMeHpj2q8mj+xGWUU1C1fv4rkP0/jjlEGEhehF\nSqW84dOMefxyYL1XyxzUph+X9Jh4wm02bFjH0qWLSUpKJiQkmMLCQiZNuvi420+cOJmJEyefsMy8\nvFxEeh1+HR0dQ35+3hHJblZWJg899ABFRUVMm3ZzrfbxNl/fpHcOMAfAGJMOxIhIJICIdAPyjDF7\njDEuYL57e7BajpVSym/WZuTy5rxNhIYE8MCVA0mMi/B3SI2KzWbjqnOSGdkvkax9xbz0yToqqxz+\nDkspdQpcLhcOh4MuXboyZsxYFi78BqfTyauvvujVY3jq1KkzN954K08//SwPP/wYf//7UzidzhPu\n0xB8/fO+HbDa43Wue1mG+9Fz5PkDQHesrhW9RWQO0Bp40hiz0DfhKqUUmB0FvDZnAwF2G/deNoAu\n7aL8HVKjZLfZuP584VBlNakmh9fmbODuS/rpOMlKnaJLekw8aWtvQ+jXbwBvv/0WIikUFRVRXV2N\n3W4nLOzYV89q08UiPj6B/Py8w69zc3OIi4s/Yv3YseMA6NChI61bx+F0Ok+4T0Pw91nrRC3DNeu2\nAI8bYy4CpgFviYhet1NK+UTm3iJe+GQdTqeLuy/pR89OMf4OqVELsNu59cI+9O3WmnXb8nhz3iac\nzoZv7VFKeV9VVRU2dza2fPkSxo+fQGlpCVVVVaSnb2Lx4u+P2H7ixMm8/PK/eOml6YcfPZNjgNNO\nG84PP3wHgDHpJCS0OSLhXrDga95//z+A1R3j4MECLrhgMosWLTzuPg3B14nmHqyW4hrtsfob16xL\n9FjXAdhjjNmLdZMexpjtIrLPvS77RAdKSNAWnpZO64CCU6sH2XuLeOHjtVRVOfif64Yxon97L0bW\nvD126xk89u8f+XnzAWJahXH35QOw2fzTW07PBUrrQP2sWbOGiIgw1q9fRXl5Mbfeeivfffcd7drF\nExYWwMUXTyQgIKBOZZ599gh++WUlv//9LQQEBPDUU09gs1Xw8ssv88QTT3DRRRN44IEHuO++ZVRX\nV/PUU08yatQo1q1bdcQ+Df2Z2nzRj6OGiJyB1Ro8XkQGAy8YY0Z7rF8PXICVLK8ArgZOAxKNMc+K\nSDvgRyDZGFN9gkO5cnKKG+x9qMYvISEKrQPqVOrBgYIynv7PGgpLK7lhQgqjNDmus7Lyav75/hp2\n7C9h/GmduOLsHj5PkvVcoLQO1N/s2e+SktLriFbg2bPfZcqUa5k16y3Gjj2Xzp27+DHC2ktIiKrT\nycenXSyMMT8CqSKyHHgBuEtErheRmlse7wA+ABYD7xtjMoAvgDEisgT4DLj9JMmxUkqdkoLi1ffK\nxAAAIABJREFUCp75II3C0kqmnJOsyXE91cwwmBgXzjc/72Teiix/h6SUqqXdu3excOHXHDiw/4jl\nV1899XDf4qaSHNeHT1uQfUhbkFs4bTFQUL96UFRWyT/eW8PevDIuGtmNSSO7NVB0LUd+UTlP/2cN\neUXlTBmXzLlDO/ns2HouUFoHFDTyFmSllGrMysqref7DtezNK+N3wzpx4Yiu/g6pWWjdKpQHpwwk\nOiKY9xduZdm6vSffSSml/EgTZKWUAiqqHLz0yVqy9xczekAiV471fX/Z5qxtbDgPXDWQiNBAZn61\nmdXpB/wdklJKHZcmyEqpFs/lcjHr63S27CpkWEobrhufoslxA+iYEMkfrhhIcFAAb8zbxI79etlb\nKdU4aYKslGrxFqft4aeN++mW2IqbJ/bGbtfkuKF0b9+K2y7sQ1W1k9fmbKCsXO+5Vko1PpogK6Va\ntKx9RcxeuIWI0EDuvKgvQYF6WmxoA5PjmTC8CwcKDjFz/mafTBurlFJ1od8ESqkWq+RQFa99tgGH\nw8Wtk/oQFx3q75BajItHdyOlcwypW3JYsGqnv8NRSqkjaIKslGqRnC4Xb83bRG5hOReO6Eq/7nH+\nDqlFCbDbuW1SH6Ijgvl40Ta27Dzo75CUUuowTZCVUi3SVz9ls3ZbHr27xjJphI517A/RkSHcPrkP\nANM/30BRaaWfI1JKKUugvwNQSilfS88u4NMl24mNCuHWSX30pjw/ks6xXDqmOx//sI1/fbGRB64c\nqJ+HUs3cyy8/x8aNG7DZbNx77wOkpPQ+vG7evM/55pv52Gw2XC4XxqTzj388xyOP/Jnu3ZNwuVwk\nJSVz330PNmiMmiArpVqUgyUVTP9iI3abjTsm96VVeLC/Q2rxzju9M1t3FZKWkcucZZlcMrq7v0NS\nSjWQtLQ17Nq1k+nTZ5CdncXTTz/J9OkzDq+fOHEyEydOPrztokULARg0aAhPPfV3n8WpCbJSqsVw\nOJ1M/3wjRaWVXHVOMj06Rvs7JAXYbDZuntiLx2euYt6KLHp0iKZ/kvYJV6qpObr112azMW3azQwe\nPPTwNqmpqxg16iwAunTpSklJMWVlZYSHh/+mvJkz3+Txx/9CVlamz0e70QRZKdVifLp4O1t2HmSI\nJHDu0I7+Dkd5CA8N4q6L+/HXd1N5Y+5GHrthGPHRYf4OS6lGIefjDyhevcqrZUYNHUbC5VedcJsN\nG9axdOlikpKSCQkJprCwkEmTLj7u9p6tv8eTl5eLSK/Dr6OjY8jPz/tNgpyevom2bdsSG9uarKxM\nsrIyeeihBygqKmLatJsZNuz0WrzL+tOb9JRSLcIvW3L4auUO2saGceOEXjpTXiPUpV0U15ybTGl5\nNa/P2UBVtdPfISnVorlcLhwOB126dGXMmLEsXPgNTqeTV1990avHOJa5c+cwYcKFAHTq1Jkbb7yV\np59+locffoy///0pqqsbdpIhbUFWSjV7Bw4e4s0vNxMcaOfOi/sRFqKnvsZq9ID2bN1VyIoN+/jw\n+61c+zvxd0hK+V3C5VedtLW3IfTrN4C3334LkRSKioqorq7GbrcTFnbsqzu16WIRH59Afn7e4de5\nuTnExcX/pqxffknlD3/4n8P7jB07DoAOHTrSunUcubk5tGuX6M23ewT9llBKNWtV1Q5e+2w9hyqq\nuemCXnRqE+nvkNQJ2Gw2po4Xduwv5vs1u+nRMZrhvdv5OyylWqSqqipqLrYtX76E8eMnUFpaQlVV\nFenpm9i/fx9jxow9vH1tulicdtpwZsz4N5MmXYwx6SQktPlNwp2bm0t4eASBgVaaumDB1+Tl5TJl\nyrXk5eVy8GAB8fEJ3n2zR9EuFkqpZu29b7eyY38JowckMqJfw7U2KO8JCQrgzov7ERocwKyvDLtz\nS/0dklItUnr6JoKCgli2bAm5ublMnnwJv/ySSlRUFBUVlYwcOabOZfbt2x+RXtxxx4289NKz3H//\nn8jPz+OZZ54+vE1eXi6xsbGHX48cOZq0tFTuuusWHn74jzz44EOHk+eGYvP1XYE+4srJKfZ3DMqP\nEhKi0Dqg1mUV8MIHv9C5TSQPTx1CcFCAv0NSdbAq/QCvz9lAYlw4j1w/lNDgun8h6rlAaR2ov9mz\n3yUlpdcRXSRmz36XKVOuZdastxg79lw6d+7ixwhrLyEhqk43nmgLslKqWdp1oITX/ruOsJBA7ry4\nrybHTdCwlDaMG9qRvXllzPra+HyYJ6Vast27d7Fw4dccOLD/iOVXXz31cN/ippIc14f2QVZKNTuH\nKqp5dc4GKqsc/P6SfrSJ/e34mo1dhaOSbQcz2VKwDVOwlT2l+6GOCWJoYCg9YrojsT2Q2CTahCc0\nudE7rji7B5l7i1i5aT/JHaMZO1iH51PKFzp06MiMGe/5Owy/0QRZKdXsvP/dVvbnl3HxWT0Y1LNh\nb+TwlmpnNZmFO9hSkIEp2EZW0Q4cLgcAgbYAEiPbEWirWyv4wYoi0nLWk5azHoCYkGh6xia5E+Ye\nxIbGeP19eFtggJ07Jvfl8Zmr+OC7DFI6x9I+PsLfYSmlmjlNkJVSzUra1lyWrdtL57aRTD2/FwcL\nGucNXk6Xk53Fu90txBlsO5hJpbMKABs2Okd1tJLZ1j1Iiu5KcEDdp8R2uVzkHsrHFGw9fJyf963h\n531rAEgIi0Nie9Aztgc9Y5OICm6cI3y0bhXK9eel8Opn63lz3iYenjqEwADtIaiUajiaICulmo3i\nskre/jqdwAAbN0/sTVBg40qiXC4XG/PSWbHnZ7Yc3M6h6kOH1yVGtKWnu2U3OaY74UGnPouczWYj\nITyOhPA4RnYYjsvlYk/pPkxBBlsKMthakMmyPStZtmclAB0iE+kTl8LZnUbSKjjqlI/vTUMkgTP6\ntOPHjfuY/2M2k0Z283dISqlmTBNkpVSz4HK5eHfBFopKK7n8rCQ6JjSe1tCaxHh+5kKyi3cCEBfa\nmkEJfa2EOLYH0SENn5DabDY6RCbSITKRsZ1G4XA62FG8+3DCvL0wi90le1m0cxmjO5zBuC5jGlWi\nfM25yaTvKGDuiiz694ija7tW/g5JKdVM6TBvqlnSYX1anp827ePfX2yiR8do/nz1YOx2m9/rwbES\n40EJ/Tiv6zl0jGrvt7iOp9JRxU97V/NN9vccrCgkyB7E6I5ncG7nsxpN94uNmfk8+2Ea7eMjeGza\nUIICT9wv2991QPmf1gEFdR/mTVuQlVJNXkFxBe8t2EJwkJ2bLuiF3e7fkRpcLheb8g1fZn5LdtGv\nifH53cbRIbLxTlYSHGAlxGe0H8aPe37mm+xFfLdjCUt3/cioRpIo9+nWmrMHd2DRmt18tiSTK8b2\n8Gs8SqnmSRNkpVST5nK5mPnVZkrLq5k6XmjrxyHdmmpifLQgeyCjO57JGe1P+02iPLrjmYzrPMav\nifIVZ/VgY2Y+3/y8g4HJ8fTs1PhH41BKNS3axUI1S3pJreX4IW0373xt6NutNX+4YsAR4/z6qh5Y\nifEW5md+S1bRDgAGJvRjQhNLjI+nylHFir2rWJC9iIMVhQTbg/yeKGfsKuTp91KJjw7liRtPO+4s\ne3ouUFoHFGgXC6VUC3Lg4CE+/C6D8JBAbpjQy+eTYDT3xLhGUEAQYzqeyZmJw1ixdxXfZH3Pwh2L\nWbJrBWM6juCczqN9nij36BjNead35qufdvDR9xlcd16KT4+vlGreNEFWSjVJTqeLGfM2UVHl4JYL\nexMbFeLT45dUlTI7/b+szdkAwMCEvpzfdVyjvPnOWzwT5eV7f2ZB1iK+3fEDy/as5OqUSxncpr9P\n47loZHfWb8vjh7Q9DOqZQL/ucT49vlKq+fJ5FwsReQ4YDjiB+4wxqz3WjQP+ClQDXxlj/uKxLhTY\nADxpjHnnJIfRLhYtnF5Sa/6+XrmDjxZlMEQSuPOivsdsPW6oerA5fwvvbvqQwspikmO6c1nypGad\nGB9PlaOKpXt+Yu62r6l0VnF6uyFc3nMyYYGhPothx/5inpq1mqjwIJ66+XQiQoOOWK/nAqV1QEHd\nu1j4dBR9ERkN9DDGnAncDLx01CYvAhcDI4HfiYjnNbNHgDyfBKqUatR255Tw6ZLttAoPYup48VnX\niipHFZ9s/YJX0t6kpKqMyUnnc8+gW1tkcgxWi/LYTqP482n30TmqIyv3pfL0zy+wvTDbZzF0bhvF\npJHdOFhSyXsLtvjsuEqp5s3X00ydA8wBMMakAzEiEgkgIt2APGPMHmOMC5jv3h53opwCfOnjeJVS\njUy1w8mb8zZT7XBy/fkptAqv+xTM9bGnZB//XP0yi3Yuo214Ag8OvYvfdTkbu61xzdbnD23DE3hw\nyF2M7zKW/PICnl/zOl9uX4DD6fDJ8ScM70z39q34adN+Vqcf8MkxlVLNm6/P7O2AHI/Xue5lx1p3\nAKi5y+UZ4H7Av4ObKqX8bt6KLLL3FzOiXzsGJSc0+PFcLheLdi7jH6tfYk/pPkZ2GM6fh91L56iO\nDX7spiTAHsCkpPO4d9BtRAe3Yn7WQp5fM53cQw1/4S/Abo1/HRxo551vDIUlFQ1+TKVU8+bvpo8T\nJbw2ABGZCqwwxmR7LldKtTyZe4uYtyKbuFYhTDmnZ4Mfr7CimNfWzuCTrV8QGhDC7f2nMUUuITjA\nN63WTVFybHcePu0PDG07kMyibP728/P8uHc1DX2/S2JcBJeelUTJoSpmfW0a/HhKqebN16NY7OHX\nFmOA9sBej3We4yJ1cC+bAHQXkQuBjkC5iOw0xnx/ogMlJER5LWjVNGkdaF4qqhy8PfNnnC4Xf7h6\nCF06xdZqv/rWg9W71/L66v9QXFHCwHa9ufO064gJi65XWS1PFP/T/jaWZv3Mm2ve5z+bPyKjJINb\nh1xNZEhEgx31qvG92JhVQFpGLuuyChh3WhdAzwVK64Cqu3qNYiEi/zDG/ElEugBtjTE/13K/M4DH\njTHjRWQw8IIxZrTH+vXABViJ8QrgamNMhsf6x4BMHcVCnYzetdz8fPDdVhas2sk5Qzpyzbm1az2u\nTz2ocFTy6da5LNuzkkB7IBcnXcCYjmf6fIzl5iLvUD6zNn3AtsIsYkKiub73lfSMbbjpoXMLD/Ho\nW9ZX0pM3nUavHm30XNDC6feBggYcxUJE4kSklfvlXPeNc/cAo0+w2xGMMT8CqSKyHHgBuEtErheR\nye5N7gA+ABYD73smx0qplsvsKODbVTtp2zqcy85KarDj7CjaxT9WvciyPSvpEJnIn4bew1mdRmhy\nfAriwlpz3+DbubD7eIoqi3nplzf4LONLqpzVDXK8+OgwpoxLprzSwYwvN+N0alcLpVTd1boFWUSu\nBgQIBlzAYOAhYL0xpmHOdPWnLcgtnLYYNB+HKqp5bMbP5BWV8/C1Q0jqUPtuDnWpB8t3r+SDLZ/h\ndDkZ22kUk7qfR1BA0Ml3VLWWVbSDtze+T86hPLpEdeK2/tOIDvH+pW+Xy8XL/11PWkYut1zUlzNS\n2nj9GKrp0O8DBQ3YgmyMmQ381RjzkDHmYeB2IBl4oG4hKqVU7f138TZyC8uZMLxLnZLj2nK6nHy+\n7Stmm/8SHhjG3QNv5tLkCzU5bgBdW3Xmz8Pu4/R2Q8gu3skzqa+wp2Sf149js9m4/jwhMiyId+Zv\n5sDBQ14/hlKqeavrKBbvisgI9/P2QJYx5h9ejkkppQDYsvMg36/ZTfv4CCaN6Ob18qscVby98X0W\nZC+iTVg8Dwy5i16tG350jJYsNDCEqb2u4MLu48kvL+C5Na+xpcD7vemiI0O4elwyFZUOZn2VrqNa\nKKXqpK4J8lbgARG50BizAmtaaKWU8rrKKgcz52/GBtxwfgpBgd4dlbKkqpSX094g9cBaukd35YGh\nd9EmPN6rx1DHZrPZOK/rOVzf+yqqHFW8kvYWK/emev04p/duy7DebdmcXcCStXu8Xr5Sqvmq6zfO\n6cCVwHUicinuWfGUUsrbPl+Wyf6CQ5w7rJPXu1bkHsrj2dRX2VaYxeA2/bln4C1EBjXc8GPq2E5r\nN5i7B95McEAw72z+kPmZ33q1pddms3HXZQMICwngo0UZ5BeVe61spVTzVtcE+XljTBVwFXAuUKvh\n3ZRSqi4y9xbx9c87SIgJ5eLR3b1bdmE2/7f6FQ6U5XJu57O4oc/V2t/Yj5Jjk3hwyJ3EhcbyZea3\n/Gfzx1R7cYSLuOgwrji7B4cqHLz7jU4gopSqnTolyMaY+e5HhzHmdqDVSXZRSqk6qXY4mTl/My4X\nTDu/FyFBAV4rOy1nAy/+8i9Kq8q4Si7hoh4TsNv8PaGoahfRlgeH3k2XqE78tG81r6+dyaFq791Y\nN3pAe3p1iWXttjxWbtrvtXKVUs3XKX0zGGO+81YgSikFMP/HbHbllB5OarzB5XLx/Y4lvLn+XWw2\nO7f3n8aoDsO9UrbyjlbBUdw7+Db6x/chvWArz6a+Rn55gVfKttlsXH9+CsFBdmYv3EpRaaVXylVK\nNV/adKKUajR25ZQwd0UWsVEhXHG2d2Zbc7qcfLz1C/6bMY9WwZHcP/gO+sb38krZyrtCAoK5pd9U\nzuo4gr2l+3lm9SvsKN7llbLbxIRxyegkSg5VMXvhFq+UqZRqvk6aIItIe/djx4YPRynVUjmdLmbO\nT8fhdDF1vBAeGnjKZZZXV/Dv9e+weNdy2ke0449Df0+nqA5eiFY1FLvNzuU9J3Np8oUUVZbw/Jrp\nbMjd7JWyxw3pSFKHVvy8+QC/bMnxSplKqeapNi3IX4hICNYYyDYRsXv+NXSASqmWYcGqnWTuLWJ4\n77YM7HHqw60VVRbzxKLnWZ+7iZTYZO4fcgexoTFeiFT5wthOo7i531RcLifT173N0t0/nnKZdruN\nG87vRWCAjXcWGMrKq7wQqVKqOapNgrsdKAXGAA6g2uNPzy5KqVO2v6CMz5ZuJyo8iCnjkk+5vANl\nOTyz+hW25WczPHEodw64kbDAMC9EqnxpYEJf7h10OxFB4XxgPuPzbV+d8igU7eMjuHBENwpLKvng\ne+9PUKKUah5Oeg3TGHMFgIi8YYy5peFDUkq1JE6Xi7fnp1NV7eSmC3oRFR58SuXtK93PS7/8m8LK\nYq7oeyGjE0Zis9m8FK3ytW7Rnfnj0Lt5de1bLMheRKWjksuSJ53SZ3r+6Z1ZnX6AZev2cnrvtvTp\n2tqLESulmoO6dJG4Q0SmishLIvKiiFzVYFEppVqMJWl7MDsPMig5nmEpbU6prD0l+3hhzb8orCzm\n8uTJXNZngibHzUB8WBx/GHwHiRFt+WHXcj7Y8hlOl7Pe5QUG2LlxQi/sNhuzvkqnvNJ74y4rpZqH\nuiTILwKTAIM15fSVIvJig0SllGoR8ovK+WhRBmEhgVz7OzmlZHZn8R5e+GU6xVUlXCWXcFanEV6M\nVPlbq+Ao7h10Gx0iE1m2+ydmp//3lJLkLu2iOO/0zuQWlvPp4u1ejFQp1RzU5TbxvsaYMR6vXxGR\npd4OSCnVMrhcLt75xlBe6eCG81OIjQqpd1nZRTt5Je1NDlWXc03K5ZzZfpgXI1WNRVRwJPcOuo1X\n0t7gx72rcLgcXJtyOQH2+k0mM3lkV9ZsyeG71F0M69WG5I56E6dSylKXFuRgz1ErRCSAuiXYSil1\n2E8b97NuWx69u8Yysn9ivcvZXpjNS7+8waHqcq7rfaUmx81cRFA4vx94K11bdebnfWuYtekDHE5H\nvcoKCgzghgkpAMycn05Vdf3KUUo1P3VJkL8EVonIcyLyHLAamNMwYSmlmrPC0kpmL9xCcJCd689L\nqXfXioyDmbyS9gaVzkpu6DOF09oN9nKkqjEKDwrj7oE3kxTdldQDa5mx8T2qnfXrR5zcMYaxQzqy\nL7+ML5ZneTdQpVSTVesE2RjzF+AuIBvIAm4zxvyjgeJSSjVj7327hdLyai4dk0RCTP2GXzP5Gbya\n9iZVzmpu6nMNQ9oO9HKUqjELCwzlzgE3kRzTnbScDby54V2q6pkkXzqmO/HRoXz10w6y9xV7OVKl\nVFNUp4k+jDE/GWNeNMa8ZIz5uaGCUko1X6kmh9XpB+jRIZpzhtRvgs7NeVt4fd0MnC4nt/a7joFt\n+nk5StUUhAaGcOeAG0mJTWZ97mb+vW4WlY66D88fGhzI9eel4HS5mDl/M9WO+t/8p5RqHnQmPKWU\nz5SWV/GfBYbAADs3TEjBXo+uFRtyNzN93UxcwK39p9Evvrf3A1VNRnBAMLf3n0afuBQ25Rumr5tJ\nhaOyzuX06daakf0T2XGghK9W7miASJVSTYkmyEopn/lg4VYKSyuZPLIriXERdd5/bc4G/r3+HWw2\nO3f0v4E+cdIAUaqmJiggiFv6XUf/+D6YggxeW/sW5dXldS7nqrE9iI4MZu7yTHbnlDRApEqppqLW\nCbKInNeQgSilmrd12/JYvmHf4fFn62rNgXW8ueE/BNgDuGvAjaS0PvUpqVXzEWQP5Oa+1zIooR8Z\nBzN5de1bHKo+VKcywkODuG68UO1wMWN+Ok7nqU1rrZRquurSgnyPiGSIyBMi0qXBIlJKNTtl5dXM\n+jqdALuNGyf0IsBet4tXP+9bw4wN7xFsD+LuATeTHJvUQJGqpizAHsANfa5maNuBbC/M5uW0Nymp\nLK1TGYOSExjeuy2Ze4tYsGpnA0WqlGrs6jKKxQRgGNYoFq+LyHwRudw9HrJSSh3Xxz9kUFBcwQVn\ndKFTm8g67fvzvjW8s+lDQgND+P2gW0iK6dowQapmIcAewPW9r+L0dkPILtrJUz+8SFlV3VqSp4xL\nplV4EJ8t3c6+/LIGilQp1ZjVdRSLAuADYDYQAzwIrBWR4Q0Qm1KqGdiUlc/itD10TIhg4pld67Rv\n6v40d3Icyj3uySFaIpfLhbOios5/rur6DXvW1Nltdq7tdTlnJA4js2Anr6y1ZlmsrajwYK79nVBV\n7WTm/M04XdrVQqmWptYz4YnIaOAG4GzgU+AmY8xmEekKfAYMapAIlVJNVnllNW9/lY7dZuPGC3oR\nGFD73+S/HFjP25s+ICQghN8PvJnOreo3JFxT4ayooCo3h6qcHKpyc63nudbz6twcnOV1v+kMm43A\n2FiC4hMIio93PyYQlJBAYHwCgdHR2OrY3aWpsNvsXJ1yKUHBdpZkr+S1tW9x14CbCQ2s3ZTmQ1Pa\nMEQSSDU5LFqzu95DEiqlmqa6TBX9N+BfwO3GmAoAEQkzxmSJyEcNEp1Sqkn77+Lt5BaWc8EZXeja\nrlWt91uXs5EZG98jyB7IXQNvokurTg0Ype+4nE4q9+ymfPt2jwQ4h6qcXBzFRcfcxxYSSlB8PIEx\nMWCrWzLrqiinKi+XQ1u3cGiL+W3ZgYEExsUfkTwHJyYSltyTgIi6jzLS2Nhtdu487TrKyitYvT+N\n19fN4M4BNxESEFyr/a/9nZCeXcAnP2yjf1JcvSe1UUo1PXVJkEuMMe8etWwJMMwY87QXY1JKNQNb\ndh7ku9RdJMaFM2lE11rvtyF3M29u+A+BtgDuHHAT3aOb7j3BNQlxmUnnkEmnbIvBWXLU8GEBAQS1\njiOkU6cjW3oTrITVHhlZ76m4azirqqjOz/u1ZfroVur9+47cwWYjpFNnwiSFcElp0gmz3W7nul5X\nUu10kJaznunr3uaO/jcQHBB00n2jI4K5+tyevDF3E29/lc6DVw085c9CKdU0nDRBFpFrgEeBLiKy\nA7ABLiAY2HeifY9T3nPAcMAJ3GeMWe2xbhzwV6Aa+MoY8xcRCQPeBtrC/2fvvuOjus6Ej/+maka9\nIiEJ0YQuSIDo3RQDxqa627FjJ3ZsJ9nsrpPdd7PJ+yabTeLNZrO7MXa8ibudrOPYsRPjAtg0gzG9\ni3oRAiRQ73X63PePGQkJhJCERm2e7+dDZm6ZuWfiR+c+99xzzyEEeEZV1fVdPa4Qovc4XB5e33Aa\nHfDY8nGYjJ17lvd05VleOfG/6HV6vp39GOnRIwNb0B6meb04i4toOnO63YTYGBtL+Oy5WMaMwZyY\nhCk+AWNMTMC7OehNJsyJSZgTk9rd7rHZcFdU4Koow15QgE09g/18Ho6CfGo2f9aSMIcqY7EqY7Fm\nZGAIHTgJs0Fv4PGsh3j1xFvkVJzk5eO/55sTvoapE0nyrMxE9p8q5VheJV8cK2LBpJReKLEQoq/p\ntE48fOAfqeI1fIly8+WzFyhUVbXTc3L6+zH/H1VVVyuKMhZ4XVXVOa22nwSWAsXAduCbwEQgTVXV\n/1IUJQ3YrKrqjWYH0MrL6ztbLDEIJSREIDHQd/687Ryf7i/gtunDeHBx58YrVqvO8buc19GAb038\nOuNiM266HIGOA03T2rQQ21QVT8OV4xljYwlVxrW0xBrj4wdMC6TX6cR+Pq/lt9nP51156K91wjx2\nHNYMBYO1f3Y/aB0DLq+bV4//gROVZxgfN5YnJjyKSX/jG6nV9Q5+9Oo+NE3jmSdmEhtpCXSxRQ+S\n84EASEiI6FLl25kW5OdUVX1aUZTRwFvt7DK/C8dbDKwDUFX1jKIo0YqihKuq2qAoykigUlXVIv9x\nNwKLVVX9n1afTwNkYEoh+rG8olo+O1DAkBgrd80f1anP5Faf900frWk8NfFrPZIcB5KztIT6fXup\n27cHV2lpy3pjTCwRs+e0tLSa4hMGTEJ8Nb3ZTOjYcYSOHQe0nzA7CvKp3vwZOqORsAnZRMyaRdjE\nbPSmzvXx7W2+yUQe4aXjv+dE5RleP/FHnhj/VQz6ju9wxESE8OCt6byx8Qy//1Tlu/dNHLD/XYUQ\nndOZPsiv+19/1APHSwIOtlqu8K87538tb7WtDGg5uyqKsgtIAVb2QDmEEAHgcnt5ff1pNA0eu2Ms\nIaYbd604X3uR3+a8jlvz8NSER8mKG9sLJe06d20N9Qf2U7d3D46LFwDQmc2ET5tB2PjxAz4hvpHr\nJsxnTtNw+BANR3z/9FYr4VOmETlrNlZlbL8bJcNkMPHUhK/xu5w3yKk4yRsn3+axrIfe20qQAAAg\nAElEQVRumCTPmziU/adLOX6+kt0nSpg7YWgvlVgI0RdumCCrqnrM/7ojAMfv6EzSZpuqqnMVRckG\n/ghkB6AsQoib9PHuCxRXNnHrlBSUtJgb7n+xroD/Ofo6bq+bb2Q9zIT4zF4oZed5bDYaDh+ift8e\nmk6fAk0DvZ7QrPFEzppN+OQp6C39s2tBoLVOmOPW3IXz8mXq9u3xtazv2kndrp0YoqOJnD6TiJmz\nCRk+vN9cPJgNJr418ev89thrHCk/juH0u3wt80H0HYwSotPp+NodY/nxa/v505ZcskbGEh3euSHj\nhBADzw37ICuKshPfQ3ntUlW1010sFEX5CVCkquor/uU8YKKqqo3+6av/1NwnWVGUf8HXwrwXKFNV\n9bJ//UlggaqqFR0cSkZ1F6KXnbtcwz8+9wXxURZe+KdbsYZ0fP19vqqAn29fS5PbztOzHmdO2rRe\nKmnHvC4X1YePUL5jJ9UHDuJ1OgEIzxhDwoL5xM+bgzk6uo9L2X9pXi91p05T/sVOKnftwe1/SNGa\nkuz7/2/+LViHtv+wYG+zuez8YsdvUCvPM3/4TP5mxqPob9DivWH3BX73lxxmZiXx/x6b0W+SfiHE\nDXXpj7UzCfKCDjZrqqp+0dmDKYoyG/hXVVWXKYoyBVjbOsFWFOU4sAIoAnYDD+HrUjFcVdXvKYqS\nCOxTVXXEDQ4lD+kFOXkoo3e5PV5+9uZBLpc38I8PTiJrRGyH+1+uL+L5Iy/T5LbxaOYDzEiaEpBy\ndSUO7AX51O74nPoDB/A2NQJgSkoicuZsImbMwpyYGJAyDmZel4umE8ep27eHxmNH0VwuACyjRhM5\nZy6Rs+agtwT2gbcbxYDNbeeFo69ysa6AOUOn85Wx93TYkuzVNP7z7SOol2r41posZoyTuOjv5Hwg\noOsP6XUmQW5+SO/qlmQdvgS5Kw/poSjKL4AFgAf4DjAFqFFV9UNFUeYBv/If531VVZ9VFMWCbwSN\nYYAFX4K94QaHkQQ5yEmF2Ls++vIC6768wPzsoXz9jnEd7lvUUMJzR16i0dXEw+PuY/bQwLUc3ygO\nNK+XhiOHqdm6uWUiDUNUNJEz+l+3gIGuve4q+tBQom6ZT/StSzDFxQfkuJ2pC5pcNn5z9GUK6guZ\nlzKLBzPu6vC/e2l1Ez95bT9mk4FnnpxJZGj/fChR+Mj5QEBgEuRsVVWPXa8lOUB9k2+WJMhBTirE\n3nO5rIGfvnmAyDAzP//GTEIt1+9aUdJYxtrDL1LvauAh5R7mpswMaNmuFweepkZqd35BzbYtuCsr\nAQjNGk/04qWEjZ/Q7x4sG2zcNdXU7NhO7fbPfTMI6nSET55C9JLbsI7J6NGLks7WBY2uJp478hKF\nDcUsSJ3DfWPWdFiOTQcu8c7WXGaMG8K31ozvsfKKnifnAwEBGOat+SE9fKNPfA3IwtfCexy4emY9\nIUQQ8Xi9vL7hNB6vxqPLlA6T49LGMp478hL1rgYeyLgz4Mlxe5wlxVRv3Uzd7l1oDgc6s5moBYuI\nXryUkOTkXi9PsDJGxxC/5i5il6+k4cB+qrds8o2EcfgQIWnDiV68lIgZM9GbbjyRR08JM4Xy95Oe\n4rkjL7Hj8m70Oj33pK+6bpK8ZGoqB86Usv90GdPHljNVSei1sgohAq9TE4VAy7jE5fj6BuuAuUCU\nqqqrAle8bpMW5CAnLQa9Y8PefN7fnsfsrCSeXHX9ESjKmspZe/hFap313DdmDQuHze2V8iUkRFBW\nVkfTyRNUb9lE04njABhj44i+dTFRtywYsFMoDyaapmHLPUvN1s00HD4EmoYhIpKohYuIXrgIY1T3\nH4rsal1Q72xg7ZGXKGksZfGw+dyVvuK6SXJxZSM/ef0AoRYjzzwxk3Br7yX0ovPkfCAgAF0smimK\nsktV1blXrfuiq32Qe4kkyEFOKsTAK6xo5Kdv3Dg5KG+qZO2RF6lx1HJP+kpuTeudKsPrcOA9fpDL\n6z7BWVIMgHVMBtGLlxI+eQo6Q+emvxa9y1VZQc22LdTu/AJvUxMYDETMmEnM4tuwjBjR5e/rTl1Q\n56xn7eGXKG0qY2naQtaMvuO6SXLzReKsrESeWpXV5fKJwJPzgYAAdLFoJVdRlKGqqhYDKIqSBOR2\n5WBCiMHB7fHyykcncXu8fG2Zct3kuMJWxXNHXqLGUctd6St6JTn2NDRQvXUzNVs3X0mwZs/pdoIl\nepcpLp6E+x4kbvVd1O3eRc3WzdTv2U39nt1YMxRiV64mdFxmQB+ejDRH8PTkp1h75EU2F2xHr9Oz\natSydo+5bMYwDqnl7D1ZyqT0eBnVQohBoivjIFuA8cAZwAuMAw5JC7Loj6TFILDe357Hhr353DJx\nKI8tb3/UikpbFWuPvESVvZo1o+/gtuGLAlomd10d1Zs/o2bbVjSHHUN4BMkr78A0fc5N3aIXfUvz\nemk6dYLqzZtoOnkC8A0TF7tyFWETsm+YKN9MXVDjqGXt4Rcpt1Vyx4glrBx1W7v7lVY18ZM39mMy\n6PnZN2YSEyETiPQncj4QEJgW5I6mmJazjhBB5uylGjbuzSch2sKDi8e0u0+VvZrnjrxMlb2aVaOW\nBTQ5dtfUUPXZRmp3fI7mdGKIiiJ2zZ1ELVhEYmq8nBgHOJ1eT9j4iYSNn4j94kUq139E45HDFD2/\nlpC04cSuXE34pMkBGXkkOiSKpyd/k7WHX2TjxS3odTqWj1x6zX6JsaE8eOsY/vCZyuvrT/G9Byah\nl+EBhRjQOt0HGUBRlEygebDKEOB5VVU7HvS0b0gLcpCTFoPAsDnc/OT1/VTW2fnhw1NJT426Zp9q\new1rD79Ihb2KFSOXtptQ9ARXVSVVGzdQt3MHmtuNMSaWmDuWEzVvPnqzb1xaiYPByXHpEpXrP6bh\n0AHQNMwpqcStWEX4tOnXJMo9EQNV9mrWHn6JSnsVq0Yt4/YRi6/ZR9M0nns/h5y8Sr6yZAxLpw27\nqWOKniP1gIDAPqS3FlgGJAHngNHAf6mq+ouuFrIXSIIc5KRCDIzX1p9i1/ESVs4Zzt3zR1+zvbO3\npG+Gs7yMqg2fULd7F3g8mOITiFm+gsjZc68ZFkziYHBzFhdRueET6vftBa8XU1IScctXETFzVstD\nmD0VA5W2atYeebHDLkO1DQ5+/Np+HC4P//L16aTEywgp/YHUAwK6niB35Z7UTH9r8VFVVacDS4HQ\nrhxMCDFwHVLL2HW8hOFJEayeO/Ka7bWOOp478hLltkqWDb+VFT3ccuwsKabktVe4+P9+QN3OLzDF\nJ5D42BOMeObfiZ6/sFfHzBX9g3loMkO/8RQjnvklkfPm4yovp+T1V7j4ox9Q+4XvzkJPibPG8PTk\nbxITEs2HeRvZUnDtHFlR4SF8/Y6xuNxXHmIVQgxMXRnFwuF/DVEURaeq6iFFUf4rEIUSQvQvNQ0O\nfv+pismo58mVmRgNba+t65z1PHfkZcqaKliatvC6T/x3h6OoiKpPPqT+wH7f7fTkFGJXrCJi+gyZ\n8U4AYB4yhKSvP07cqtW+bjdffkHpH96g8pOP8Nx/N4ZJM9EZu3K6a1+8NdbXJ/nIi3xwbj16dNeM\nzDIlI4F5E4fyZU4xH355gXsWXHunRQjR/3Wli8VLwDEgDZgGqMAcVVUnB6543SZdLIKc3FLrOZqm\nsfa9HI6fr+ThpRksnpraZntXJlboCmdpKZUfr/PdPtc0Qoal+R7Imjyl04mxxEFwctdUU/XpRmq/\n2I7mdGKMiyNu5WoiZ8/tkUS5rKnCP/FNHfeOWc2iYfPabLc53PzrG/upqLXzzw9NIWOYPM/el6Qe\nEBDYPsg6IAaoAR4EEoH3VFW93NVC9gJJkIOcVIg95/PDl/nfTWcZPzKW793fdlitBmcjzx15iaLG\nEhYNm9fh1Lyd5aoop/KTj3x9jL1ezKnDiF9zF2GTJnf5uyUOgpu7thbb9s2UbPwUze3GlDCEuNVr\niJg5+6bvPpT6Z4esc9Zzf8adLEid02Z77uUafvnHw8RFWvjp4zOwhtx8Yi66R+oBAYGdKCQUX2Kc\nhW9c5ONAVVcOJoQYWIorG3l32znCLEYeWz6uTYJa72zgN0dfoaixhAWpc286OXZVVVK1/mNqv9wJ\nHg/mocnErbmT8CnTpCuF6BZjVBSjnngMy/zFVG34mNovdlDy2itUrv+YuNV3EjGt+910EkMTWrpb\n/PnsOnTA/FZJ8pjUaJbPGs76Pfn8aUsuj6/ojwM+CSGupysJ8vtAObAb0AG3ACuBVQEolxCij7k9\nXl795BROt5cnVma2mfygxlHLb468QklTGfNTZnPfmNXdTo7dNTVUbfjEdzvc7caUmOhLXqbPlMRY\n9AhTTAyJDz9K7O0rqFr/EbW7vqTk5RepWv8JcavvJHzK1G7Fb1LYEJ6e/E2eO/wS755dh9PrYkna\ngpbta+aN5MT5Kr48Xkx2ehxTlSE9+bOEEAHUlS4Wu1RVnXvVui9kJj3RH8kttZu3bud5Ptp1kdlZ\nSTy5KrNlfaWtiuePvEyFvYpbh93C3ekru5VcuOvrqN64gZrt29CcTkzxCcSuWk3krDktQ3TdLIkD\n0V4MOMvLqPr4Q+r27Pb1b08bTtyauwibeOOZ+dpT2ljG80dfocZRy/KRS1k+YknL9xRVNPLTNw8Q\nYjLws2/MIDpcZtnrbVIPCAjsMG+5iqIMbV5QFCUJyO3KwYQQA0NeYS2f7M4nLtLCw0szWtaXNpXz\nrH8SkDtGLOlWcuxpaKDir+9z4Qf/RPWmTzGEhzPkka8z4pl/J2ruLT2WHAtxPeaEISQ9/iQjfv4L\nImbMwnGpgKLfrOXSv/+cxpMn6MoEWgCJYUP43pRvE2eJZcOFzazL29DyHcnxYdy3cDQNNhdvbDjT\n5e8WQvSNG7YgK4qyE1+fYwswHjjjXx4LHJIWZNEfSYtB99mdbv71jQOUV9v4/kOTUdJiAChqKOH5\noy9T72zgztHLWTp8YZe+19PUSPXmTdRs2YTXZsMQFU3sipVE3bIgYGMYSxyIzsSAo/AylR+to+HQ\nQQCsYzKIW3MXoWO71m+4xlHL80deprSp3Nf1KGMNep0er6bx7J+PcfJCFY/clsGiKak3/jLRY6Qe\nEBCYh/R+1M2yCCEGoHe3naOs2sbtM9NakuOCusu8cPRVGt1N7T6x3xGPzUbNlk1Ub/rUlxhHRJBw\n/4NELby1ZUpoIfpSSEoqyd/+W+wF+VR+tI7Go0e4/F//gTVDIe7OuwnNUDr1PdEhUXx3yrd44eir\nfFG4B4fHycNj78WgN/D48nH8y2v7eHfbOcYOj2FonMyyJ0R/1pU+yAbgIWA6vhbkvaqq/imAZbsZ\n0oIc5KTFoHuOnqvg+fdzGDYknB89Og2TUU9ezUV+e+x1HB4HD4+7j9lDp3Xqu7x2G9Vbt1D92ad4\nmxrRh4cTe/tyohctRh/SO/0wJQ5Ed2LAfvEClR9+QOPxHABCx2USt/ourGPGdOrzja4m/ufoa+TX\nX2LykIl8PfNBjHojB86U8bt1Jxg5NIIffnXqNRPuiMCQekBAYId5ex4YAmzHN4rF/YqizFJV9emu\nHFAI0T/VNTp5c8NpjAYdT67KxGTUc6Yql5dy3sSteXgs6ytMTZx0w+/x2u3UfL6Vqs824m1oQB8a\nRvzd9xJ962L0Fmsv/BIhbo5lxEhSnv4HbOfzqPzwA5pOnqDp9ClCs8YTt/pOrKPTO/x8mCmUv5v8\nJL879gZHynJweVw8Mf6rTB87hKNZSew5WcInuy9y5y2jeukXCSG6qisJ8nhVVRe0Wn7B3z9ZCDHA\naZrGmxvPUNfk4oFb00lNCOd4xSlePfEWaBpPTXiUCfGZHX6H1+GgZvs2qj/dgKe+Hn1oKHF33k30\n4qUYrJIYi4HHOmo0qd/7P9hyc6n8yJ8onzxB6PiJxK+5E8vI6ye4VqOFv530DV4+/gdOVJ7mdzlv\n8M2JX+fhpRmcvVTNJ7vzGT8qjvSUqF78RUKIzurK/R2zoigt+/u7XMjUQEIMApsPXOLouQrGDY9h\n6fRhHC7L4eXjf0CPjm9nP95hcux1Oqne/BkXfvhPVLz3LprbTeyqNYz85X8St3K1JMdiwLOOGUPq\nP36f1O//EGuGQtOJHAr+7WcU/mYt9oL8637ObDDzzYlfZ2J8Fmr1OV44+io6o4snVmaiaRovfXiC\n+iZnL/4SIURndaUP8o+Au4Ad/lWLgHdUVf2PAJXtZkgf5CAnfc467+ylGn719hEiQk385LHpnKk/\nzlun3yPEYObb2Y+THj2y3c95nU5qd+6gasN6PLU16EIsxCxdSsySZRjCw3v5V7RP4kAEIgaazpym\n8sMPsOWeBSBs8hTiVq3Bkja83f09Xg+/P/UOh8qOkRaRwncmPcG2/WWs23mBrJGxfO++bPT6m5ui\nXVyf1AMCut4HudMJMoCiKLOAmVx5SG9/14rXayRBDnJSIXZOTYODn75xgPomF99/aDIlulO8e3Yd\nYcZQvjPpGwyPHHbNZzw2G7Xbt1G96TM89XXoQkKIvnUJsbfdjiEiog9+xfVJHIhAxYCmaTSdOknl\nhx9gP58HQNiEicQuX9Xuw3xezcvbZ/7CnuIDJIcl8Z3sJ3jz4wvk5FWyas4I7pov/ZEDReoBAQF8\nSE9RlMdUVX0D2NvlUgkh+h23x8uL605Q2+jkwVvTKdCO8UHueiJM4fzd5CdJCR/aZn9PQwPVWzZR\ns20L3qYm9FYrsctXEr3kNoyRkX30K4ToGzqdjrCs8YRmZtF08gRV6z+m8XgOjcdzsGYoxK5YRWhm\nVstEOnqdnofG3oPZYGbH5V08d+RFvnHbYxS93cjHuy8yKjmS7PT4Pv5VQohmXeli8THwVVVVawNb\npB4hLchBTloMbuydrblsOnCJqWPjGZJ5ge2XdxEdEsXfT36KxNCElv3cNdVUf/YpNV9sR3M4MIRH\nEL30NqIX3YohtH+P5SpxIHozBprOqlRtWE/TCd/wcCEjRhK7fCXhkyaj0/se4dE0jY/Of8qm/M+J\nMkeyJuV+Xnu/CLNRz788Np0h0dJnv6dJPSAggF0sFEXZAkwFVKDlqQKZSU/0R1Ihdqx5PNakeDPJ\nU89yuvoMyWFJfGviY8RZfZODOMvLqN64gbrdX6K53RhjYohZdodv5rteGsf4ZkkciL6IAXtBPlXr\nP6bh8CHQNMxDk4ldvoKI6TPRGX03brcWfMEH59Zj0huZGXYHm7Y6SRsSzv99ZCpmk0y33pOkHhAQ\n2AR5QXvrVVXd0d76PiYJcpCTCvH6iioa+fkfDoLRztBpJym1lzAuNoNvjH8Yq9GKo/AyVRvWU79/\nL2gapiGJxN6xnIhZcwI2JXSgSByIvowBZ3ERVRs3ULdvD3g8GOPjib19OZFz56E3mTlWfoI3Tv4J\nt9fNMPcM1MMxzJswlMeWj23pmiFuntQDAgKQICuKEolvuumxwE7gWVVV3d0uYe+QBDnISYXYPrvT\nzc9/f5ASWwmxE3No8jYwN3kmD2TcifNiPlUbP6HxyGEAzCmpxK5YScTU6egMA7NFS+JA9IcYcFVW\nUPXpRup27kBzuzFERRFz2+1EL1jIJWcFL+a8SZ2zHmt9OlWnR/G128exYFJKn5Z5MOkPMSD6XiAS\n5LeAIuAL4G6gUFXVH3e3gIqi/BqYBXiB76qqerDVtiXAvwFuYKOqqs/41/8KmAcYgF+qqvrBDQ4j\nCXKQkwrxWpqm8eKHJzlUfAJrRg6azsOdI5cxo9RKzedbseedA8AyahSxy1cRlj1pwLdiSRyI/hQD\n7toaqjdvonb7Nrx2O3qLhci5t6CbO52Xiz+hqLEE6obgysvmhw/NZORQefi1J/SnGBB9JxCjWIxQ\nVfWrAIqibAS2dqdg/s/PB9JVVZ2jKMpY4HVgTqtdngOWAsXADkVR3geSgEz/Z2KBI8CNEmQhxFU2\nH7zM4aoDhGScIdyp4+Ha0VjXf0RJbQ3odIRNzCZm6TKsY8cN+MRYiP7IGBVNwr33E3vHCmo+30rN\n9m3UbN0MWzfzyPjx7BqZwJcRpRgy9vLCx3p++sh8wq0Dq1uTEINFZxJkV/MbVVU9iqJ0fuDkay0G\n1vm/64yiKNGKooSrqtqgKMpIoFJV1SIARVE2+Pf/LbDP//kaIFRRFJ2qqjdTDiGCypmCKv567iNS\nI84xdZ+LjHwbeErQrFail9xG9KLFmBMT+7qYQgQFQ1gYcStXE3v7choOH6J662bsJ04w9QRkxYSx\nZ5SNU6nb+M0GHf981yKZRESIPtCZBPnqRPRmEtMk4GCr5Qr/unP+1/JW28qAUf5E2OZf9wSwQZJj\nITqvtKqazz55lgcLikmu8D0+YEpKImbxUiJnz0FvkWGlhOgLOqORiBkziZgxE/vFi9Rs24xu/z4W\nHXIz51gjp0a+zx/X1/PIqjV9XVQhgk5nEuQ5iqIUtFoe4l/WAZqqqmk3cfyOLovbbFMUZQ3wGHDb\nTRxPiKDhrq2lZNunlG3dzB12X2JsGT+euKW3Ezous2VcViFE37OMGEHS408Sf+8D1H6xnfItnzE5\ntxEt9wP2H91H1l1f8U08In+3QvSKziTISg8erwhfS3GzZHz9jZu3tZ66K8W/DkVRlgE/BJapqtqp\nnvYJCf1rylvR+4IxBjSPh9oTJynb9jnlX+4CtweTScfpMck88PT3iRh27dTRg10wxoFoa0DFQEIE\nQ0c/TMYjD7Dz/T9TtuljkvOLKFz731hSkklauoSEBfMxx8b0dUkHlAEVA6Jf6PQ4yD1BUZTZwL+q\nqrpMUZQpwNrWE40oinIcWIEvMd4NPISvq8VOYLGqqhWdPJSMYhHkgu2pZUdRIXV7dlO/dw/u6ioA\nqiOMHFUsXIicwr/c/QihluB72CfY4kBca6DHwMbDZ9h38n+ZVFDJ2HwHeq8GOh2hWeOJnD2X8EmT\nB8zEPX1loMeA6BkBmyikpyiK8gtgAeABvgNMAWpUVf1QUZR5wK/w9XN+X1XVZxVFeRL4CXAWf7cO\n4FFVVS93cBhJkINcMFSI7vo66vfvo27PbhwXLwCgt1ipGjuUzfFVFMaF4C2YxI9XryIlIbyPS9s3\ngiEORMcGQwy8suEYhxyfEmYpZ0qxkRmFJrT8SwDoLRbCp00ncvZcrGMypAtGOwZDDIib1+8T5F4i\nCXKQG6wVotflojHnKHV7dtN4PAc8HtDrCRs/Af20bN4xnOJcYwF6VyhNZ7N5avEcZmYG7+gUgzUO\nROcNhhhwujz821sHKDYfxjT0Ika9kfui56Gcb/LdNaqqBMAYF0fk7DlEzpqLOSnpBt8aPAZDDIib\nJwmyjyTIQW4wVYiapmE/n0fd7l3UH9iPt6kRgJC04UTOnkPEjFmcdF7irdPv0eS2YbUNo+pUBrdN\nGcWDi8f0cen71mCKA9E9gyUGymtsPPOHgzSZi4gcewq710Z2fBYPKfegv3DJVz8cOojmsANgGTWa\nyNlziZg+A0N4cN5BajZYYkDcHEmQfSRBDnIDvULU3G6azqo0HjtK47GjuCp8IyAaoqKInDWbyNlz\nCUkdhsvj4oO89ey4vBuT3khs/VQunoxmxrhEnlqdhT7IJ/wY6HEgbt5gioELxXX86u0jeAw2Rsw8\nR6GtgJiQaB7LeojR0SPwOhw0HDlE3Z7dNJ06CZqGzmjEqowlPHsSYdmTMMXF9/XP6HWDKQZE90mC\n7CMJcpAbiBWip76exuM5NBw7QtPJE3jtvpYgvcVCWPYkImfP9Q3PZjAAUNpYxusn3+ZyQxFDwxKJ\nq57NgaN2MkfE8PS92ZiM0hdxIMaB6FmDLQZOXqhi7XvHMJt0zFvSyK7yL9DpdKwYuZTbhi9Cr/P9\n3buqq6nft4f6fXtxXLoyUqs5JbUlWbaMHBUUfZYHWwyI7pEE2UcS5CA3ECpETdNwFhXReOwIDTnH\nsOedA//foykhgbDsSYRNnERohoLO2HZExn3Fh3jn7Ac4PU7mJs/EVDqejXsKGZ4Uwfe/MhlrSGdG\ncBz8BkIciMAajDGw91QJL390iqgwMw/fFc8H+X+hxlGLEpPO1zIfJCokss3+rqpKGo8do+HYUWxn\nTqG5feOiGyIiCJuQ7atrsrIG7aRBgzEGRNdJguwjCXKQ668VotflwtbcdSLnWEvXCXQ6rOljCJs4\nibDsbMxDk9G10z3C7rbz7tl17C85jMVg4aGx91B9KZa3t+QyJMbK//3qVCLDzL38q/qv/hoHovcM\n1hjYfOASf9qaS2KMlb9/YBwfFqzjeMUpwk1hPJr5IFlx7U9h4HU4aDp1koZjR2nMOYqnrg6gpStG\nWPYkwidmY4pP6M2fE1CDNQZE10iC7CMJcpDrLxWip7ERW14u9nPnsOWexX7xAprLBYDeaiU0a4Lv\nduf4CRgiOh7I/lJ9Ia+f+CNltgqGRw7j8ayHOX/RxUsfniQyzMwPH5nKkOjB2QLUXf0lDkTfGcwx\n8Jcdeazfk8+IpAj+z4OT2Fe+j3Xn1uPWPCxJW8CqUcsw6q9/N0nzenHkX/Qny8dwFOS3bDPGxWFN\nH4M1PQNr+hjMKSkDtjvGYI4B0XmSIPtIghzk+qJC1DQNV0U59nO52HJzsZ3LxVlUeGUHnY6Q1GEt\nD8xYx2Rc03Xiet+7/fKua058akEda/98DLNJzz8/NIW0RJkp6mpyYhSDOQY0TeONjWf4MqeYrBEx\nPH1fNkVNRbxx4u1WF9IPEW+N69T3uaqqaMw5SuOJ49jO5eJtaGjZprdasYxO9yfNY7CMHDVgJigZ\nzDEgOk8SZB9JkINcb1SImtuN4/IlbLlnsZ3zJcSe2tqW7TqzGcuo0VjH+FpgLKNGY7B2rYW30lbF\nn8+u40TlmTa3TvNL6vnl24fxeLz8w/2TGDtcpp1tj5wYxWCPAY/Xywt/Oc6xvEpmZiby5KpMnB4H\n76jrOFDq64p1d/oKZidPb3mArzM0TcNVUuyv285hO3cWV2nplR0MBkKGpfnrN9C3AQcAACAASURB\nVF/ibIyKDsAvvHmDPQZE50iC7CMJcpDryQpR83pxlZXhKLyMs6gQZ1EhjsJCnKUlvok6/AxR0VjH\njGlpYQlJHdapFuL2uDwuthTs4LP8bbi87jYP35RWN/Hv/3uI+iYX375zPNPGDumR3zkYyYlRBEMM\nOFwe/vvdo5y7XMuSaal8ZfEYdDod+4oP8e7ZD3B4nAyPHMYDGXcyPHJYt4/jrqvDdi7Xd5fsXC72\n/Itt68CICMzJKYSkpPhfUzEnp2AIC+uBX9l9wRAD4sYkQfaRBDnIdadC1LxeXJUVOAtbJcFFl3EW\nF7c89d1MF2IhJCXZ14Li76NnjI9v98G6rjpRcZr3cj+iwlZJpDmCu9JXMD1xMjqdjtoGB7946xDl\nNXYeWaawaHLKTR9vMJMTowiWGGiwufiPPx6msKKRexaMYsXsEQDUOGr54Nx6DpYeRYeOuckzWDX6\ndsJNN5+0ep1O7Bcv+BLm83k4Cy/jqqhoGY2nmSEq+krSnJyC2f++q3fUuitYYkB0TBJkH0mQg1x7\nFaLmduOurcFVWYm7qhJ3VRWuqircVZW4KitxlZehOZ1tPqMzmzEPTW5TqYekpGKMje2RZLi1ClsV\n7+d+xPGKU+h1ehamzmX5yKVYjRYAmuxu/uPtw1wqa2D13BHcecuoHj3+YCQnRhFMMVBVZ+cXbx2i\nqs7BY8vHcsvE5JZtZ6vP8e7ZDylpLCXMGMrq0bczJ3lGl7pddIbX4cBZXORvYChseW2eDrs1Y0ws\npvh4jLGxvvdxcRhj4zDFxmGMjUUfGtoj9WwwxYC4PkmQfSRBDiJehwNPYwOehga8jY14GhqweB3U\nFBT5kl9/EuyuqbmmZaOZLsSCKSHhmluDpvj4gD+57fS42Fywnc35n+PyukmPHskDGXeRHJ7Uso/L\n7eHZPx/jTEENCyen8MhtGT2eoA9GcmIUwRYDRRWN/Ptbh7A5PPzt3ROYNObKzHker4ftl3ex/sIm\nHB4naRGpPKjcdVPdLjrLY7P5uqgVFuLwd1VzFhfjrqnuuF6Oi22TNBtjYjCER2AID8cQFo4hPNyX\nSHdQTwdbDIj2SYLsIwnyAKB5vWhOJ16HA6/TgeZw4HU4rqxz+NfZbXgaGlqSYI8/Cfb6l5uHTbsu\ngwFjdLS/go3ztVLExGKMi73SUmHtmZaKrjpecYr3z35Ehb2KKHMEd6WvZFripDZl8Xo1XvzwBAfV\ncqZmJPDtO8ej10ty3BlyYhTBGAN5RbX855+OoGnwTw9OJj01qs32q7tdzEmezupRdxBu7v2+wprH\ng7umxt+Y0XxnrxJ3ZSXu6ipclVV4mxo7/hKdDn1oaJuk2RAWjj48HENYGFGJcTQ6NfQhIehDQtCF\nhKA3h6C3hKAzh1xZ381nRsTAIAkyUL5jp1ZXb+vrYlyrO/9fa9ddaLPY9r+j5t+m+Y6pgda8TvO2\n2aY1l0sDvF40zQteDTQvmtfr28frhavet7x6PGgeN5rbjeb2vafVe83j8W9zX9mvVSLcXXqrtU0F\naPC/6v2VY0xKIjZTKMbYOIxRUf1u/M4KWyXvnf2IE5Wn2+1O0cyrabz1mcr2o0Uow6L5hweyMRkN\nfVTqgScYkyPRVrDGQE5eJc+/n4PFbOD7D01udxjIs9V5/PnsOor93S5Wjb6duQHodnGzvHab/05g\nFe7qKjwNja0aTBrwXtWA0vrBwS4xGNCbzS0JtM5oQGcwojP6/mEwoDMYWpZ92wz+bUZ0Bj06nR70\netDpfOedq97rdDrfa/M2dKDXoQPQ6XzLurbvde2s8/8Pvo+1yvvapIDXW99J/fQupc5gIGz8hC7P\n/CgJMrBrzT2D70cNVM2VicHoq1hMRl/F03zFbg5BH2L2X9VbrlzJm80t7/UWS0viawjzJcI3utLv\nrydFp8fF5vzP2VSwHbfXzZjoUdyfcWeb7hTNXG4vr60/xf7TZQwbEs4/PzSFUIu0cHRFf40D0XuC\nOQZ2nyjm1U9OYw0x8Hd3T2x3OMjmbhcbLmzG7nGQFpHCA8pdjIhM64MS3zxN0/Da7f6k2ZdIhxm8\n1FbUXLkr2dJI0+q9w4nX2erOpdMJbRqA3N1r5BIBEX/P/cTesbxLn5EEGSj5bJNWX+/o62K0rxsX\nZLqOrgLbXOFdvZ/Od2XZfLWpv/qKtNU2Hb4rWp3edzXrf6/zX/22XPG2vhLW631XzUbjlSvtVlfX\nGAx91k+2v50U7W47u4r2s+3STmoctUSZI7g7fSVTr+pO0czmcPPCX49zOr+aMalR/N09Ewm3mvqg\n5ANbf4sD0fuCPQb2nSrl1U9OodPBU6uyrjssZK2jjg/OredA6REAsuOzWDp8ISOjhvdmcQOip2JA\n83qvJMvt3D3VPB7/HVbfXVjfXVmtzV3XNndi/dta7vpq2pW7va3uADdv9+VrzetaStXqrdb+6qvv\nPnfqx3b9I71FZzAQPmkyhvDwLn1OEmQf6YMc5PrLSbHOWc/2S7v4onAPNrcNs8HMgpQ53D7iVixX\ndadoVtvg4Nk/H6OgrIHJY+L55uoszCbpVtEd/SUORN+RGIBTF6t44a/HcTg9PLQ0g8VTU6+7b251\nHh/kbSC/7hIA6dEjWZq2kKy4sQP2wWCJAQGSIDeTBDnI9XWFWNZUwdaCHewtOYTb6ybcFMbC1LnM\nT51DmCn0up8rqWri1+8epaLWzoJJyXz1tgwM/az/9EDS13Eg+p7EgE9+ST3PvneMukYnK2YP5+75\no66b8GqaRm7NeTbnb+dUlQpAclgSS9IWMC1xEgb9wLpglxgQIAlyM0mQg1xfVYgFdZfZVLCdo2XH\n0dCIs8SyJG0+s4ZOw2wwd/jZ80V1rH3vGA02F3fOG8mquSMGbItNfyEnRiExcEVZjY1n3z1KabWN\nuROS+NrtYzEaOr4AL2woZnP+dg6VHcOreYkJiebWtFuYM3QGFmNIL5X85kgMCJAEuZkkyEGuNytE\nTdM4U5XLpoLtnK0+B8Cw8GSWDF/I5IQJnWptycmr5LfrjuNye3l0mcKCSTJDXk+QE6OQGGirrsnJ\nc+8d40JxPRNGxfE3d44nxHzjOqrSVsW2SzvZXbQfp9dFqNHKgtQ5LEidS4S5a31Be5vEgABJkJtJ\nghzkeqNC9Hg9HCnLYXPBDi43FAGgxKSzdPhCxsaM6XTr767jxbyx4QwGg45vrcli8piEQBY7qMiJ\nUUgMXMvudPO7dSc5fr6SkUMjefq+iUSGdnyHq1mDs5EdhbvZcXkXja4mTHojs4dOZ3HafOKtcQEu\nefdIDAiQBLmZJMhBLlAVolfzcqG2gENlxzhSlkOdsx4dOiYPmcDStIWkRV7/4ZeraZrGhr35/GXH\necIsRv7+3omMSY3u8TIHMzkxComB9rk9Xn6/8Qy7TpSQGGPlHx6YREJ058eVdXic7Ck+wLaCL6i0\nV6NDR3r0SKYmZjMpYUK/alWWGBAgCXIzSZCDXE9WiJqmcbHuEofLjnG4LIcaRy0AYcZQpiZms2jY\nLQwJjb/Bt7Tl9Wr8aWsuWw9dJjYyhO/dP4mU+N6fxWqwkxOjkBi4Pk3T+MuO82zYm09UmJnv3pfN\n8KRrJxTpiMfr4XBZDl8U7uF87UUA9Do9GdGjmZqYTXbC+A4fTO4NEgMCJEFuJglykLvZClHTNC7V\nF3LInxRX2asBsBqtZCdkMXVINkpMeree5na5PbzyyWkOnikjJSGM792XTWxk+0O+iZsjJ0YhMXBj\nmw9e4p0tuYSYDfzt3RPIHBHbre+pttdwuCyHQ2XHWoaJ0+v0jI0dw9Qh2WQnZGE1dm32s54gMSBA\nEuRmkiAHue5UiJqmUdhQ3FLBV9gqAbAYQpgQn8XUxImMi83AqO/+bHZNdjcv/DWHMwU1ZKRG8Xf3\nTiTMIhOABIqcGIXEQOfsP+2bUETT4ImVmczMTLyp76u0VbXUpZfqCwEw6gyMi8tgypBsJsZnXncs\n+J4mMSBAEuRmkiAHuc5WiPXOBs7XXiSv9iInKk5T2lQOgNlgZkLcOKYmZpMZq2Ay3HwSm19Sz8sf\nn6S4sompGQk8tToTk3FgjSc60MiJUUgMdN7p/Gpe+GsONoeH5bOGc+ctI284DFxnlDWVc7gsh8Nl\nORQ2FANg0hvJjFUYEzOa0dEjSAkbGrDxlSUGBEiC3EwS5CDXXoWoaRpltgrO1/gS4rzaC5Q1VbRs\nN+mNjI8bx5TEbMbHjb3huMWd5fVqbNyXz7qdF/B4NW6bPoz7F6Wj18sYx4EmJ0YhMdA1BaX1vPDX\n41TU2kkbEs6TqzJJSei5B+5KGks5VJbD4dJjlDSVtawPMZgZGTmcUVHDGR09khGRw3qshVliQIAk\nyM0kQQ5yCQkRFJdWc6m+kLzaiy1JcYOrsWUfiyGEkVHDGR01glFRIxgRlUZIDyXFzcpqbLz6ySnO\nXa4lKtzMN5aPY/yo/jkU0mAkJ0YhMdB1Noebd7bmsjOnGKNBz70LR7NkWir6Hpy4SNM0KmxV/jt4\nF8irudgmYdahIzUiuaV+Hh09guiQqG4dS2JAwABIkBVF+TUwC/AC31VV9WCrbUuAfwPcwEZVVZ/x\nrx8PrAN+rarqbztxGEmQg4hX81LjqKWsqYKypnLKmioodZSSW3kBl9fdsl90SBSjo0YwOnoko6JG\nkBKehF4XmGmcNU1jZ04xf9qai8PpYdrYITy6TCHcKv2Ne5OcGIXEQPcdOVvOm5+eob7JxbjhMXxj\nxbiAPlDc4GrkQm0+ef4GjYK6S7g1T8v2OEsMI6OGkxQ6hCGh8QwJTSDBGn/DGf0kBgR0PUHu/tNG\n3aAoynwgXVXVOYqijAVeB+a02uU5YClQDOxQFOV9oAB4HtjSm2UV/YumaTS6miizlVPaKhEuayqn\n3FbRJhEGX+tDcnhSm9aHWEtMr5S1rtHJ7z89w5HcCqwhRp5clcmszESZNloIMaBMzkhgVEoUb244\nzbG8Sn782n4euS2DWVlJATleuCmMCfGZTIjPBMDlcVFQX9jSwnyhNp+DpUev+VyUObIlYR4SGk9i\naAJDrPHEWWNv6qFqEdx6tQVZUZSfAvmqqr7uXz4FzFBVtUFRlJHA71VVne/f9gOgHvgdYAJ+AJRL\nC/LgomkaNredemc99a5G6p0N/n9Xln2tw+U0uW3XfD7EYPZVitb4NhVkVtooGmvc7RwxsI7mVvDm\nxtPUNbkYmxbNN1ZkEhclQ7j1FWk5EhIDN6/ljtiWXBwuDzPGDeGRZUqvj8Dj1bxU2qops7VtJCmz\nVVBtr0GjbT6j1+mJs8SQHDUEC6GEm8OIMIUTYQ4nwhxBhH853ByOSRLpQa9ftyADScDBVssV/nXn\n/K/lrbaVAaNUVfUCDkVReq2Q4sY0TcOteXB6nDg8DhweZ5v3jmvW+5ZtbjsNzsY2CbCn1S209hh0\nBuKtcYyOHulrHbAmtCTDkeaIdltmQ01WGum9k6Ld6eadref44lgRRoOeB29NZ8n0YT3aZ08IIfqC\nTqdjfnYySlo0r35yiv2ny8i9XMvjK8aR1c0xk7tDr9OTEBpHQmgcWXFtcwKnx0WFrfJKNzub77W8\nqYJjJadv+N1Wo7UlYY4wRxBmCiXEYCbEEOJ/9b03X+d9iMGM2WAKWLc90fv6+pKpo+yh25nFKwff\nxmZ3dffjndONlveOP9H+1iuHuXJt3PJOo80Vc/N7TQMNL5qmoaG1vHo1rWW9t3l9y3svHs2Dx+t7\ndWsevF7fq8fr8W3zv3drHryat8u/vzWzwUyEKZxhESm+q/mWq/rwa5bDTKH9utI5d7mWVz45SXmN\nnWH+p75Te/CpbyGE6A8SY0L5wcNT2LC3gI++vMB/v3OUJVNTuXfhaMymvh2y0mwwkRyeRHL4td0/\nImLMXCwqoc7ZQIOrodWdygbqr1oub6q8piW6K3ToMOgNGHR6jDojer3v1aDT+9cbMOgNGHUG9Drf\nfjqdDr1Ojw4dOp2u5VWPDl3z9lbbQIdOh//9leP637Qst9naYUbV/sZuJWG90Chk0BlYkDqHxNCE\ngB6ntxPkInwtxc2S8fU3bt42tNW2FP+6Ltuct7NbhQt2ev8fsNH/x+t7b8RkDLlmXfN+FmNIy7+Q\nVu+v/hdiNGMxhmA1Woi0RNzwoYqekJDQtSlTu8rl9vLOZpX3t55FA+69dQwPLVNkbON+JtBxIPo/\niYGe9fiaCcyfMoz/fvsQWw5dRr1cwz88NJX01Oi+Ltp1jU0b3qn9vF6v/y5nI3a3A4fbgb3lnxO7\n296y7HA7W7333TH1eD24vf7GJK+7pZHJ7fXg8rj867wt20X3jExIZvzwUQE9Rm/3QZ4N/KuqqssU\nRZkCrG3uc+zffhxYgS8x3g08pKrqOf+2nwAVqqr+z42Oc7muWKuuarzRbj2g61dKHV/E3egq7so1\n4ZVd279K7PBqFB36VutbtvfjVtquCmS/Q6+mcUgtZ93O8xRXNhEfZeGJlZlkDOu/J4dgJf1PhcRA\n4DhdHt7fnseWQ5cx6HXckp3MytnDAzrSRXf01xhofZfX2+pur6Z5/Xd8278D7Fvf8i1t7iP7vrfd\ne87XK0T7q7v3i7r1qa7S6wwkWOO6/OD7QBjm7RfAAsADfAeYAtSoqvqhoijzgF/h+3/5fVVVn/Un\n0v8NDAdcQCFwt6qqNR0cRh7SC3KBqBA1TePw2Qo+/PI8l8sb0et0zM8eyn2L0rGG9HVvJdGe/npi\nFL1HYiDwTl6o4q1NKqXVNowGHQuyU1g+ezgxEYG/U9gZEgMCBkCC3EskQQ5yPVkhaprGsbxK1u08\nT0FpAzodzMpMYvXcESTGhvbIMURgyIlRSAz0Do/Xy54TpXy06wIVtXZMRj0LJ/kS5aiwnp2Aqask\nBgT0/1EshBgwNE3jxIUq1u08z4XienTAzMxEVs8dwdC4sL4unhBC9BsGvZ55E4cyKyuR3SdK+HjX\nBTYfvMSOo4XcOjWV22emERnat4myEF0hCbIQV9E0jVP51azbeZ68wjoApikJrJk3khQZnUIIIa7L\naNAzPzuZOeOT2JlTzCe7L/LpvgI+P1zIkmmpLJuRJjOKigFBEmQhWlELqvlg5wXOXvJ1cZ88Jp41\n80aSlihPwQshRGcZDXoWTU5h3oQkdhwtYv2efNbvyWfrocssnTaMZTOGEdrLE40I0RWSIIug5/F6\nOX2xmo37CjidXw1A9ug41twykhFJkX1cOiGEGLhMRgNLpg1jfnYy248UsmFvPh/vvsiWQ5e5bfow\nbpk4tN+NeiEEyEN6YpC60UMZmqaRV1jH3lMlHDxTRl2Tb2KZ8SNjWXPLSEYnR/VWUUUAycM5QmKg\nf3E4PWw7cpmNewtosPnq3Yxh0czMTGSakkBEAPopSwwIkFEsmkmCHOTaqxA1TeNSWQP7Tpey/1QZ\nlXV2AMKtJqaPG8Kc8UmSGA8ycmIUEgP9k83hZu+pUvadKm3p0mbQ68gcEcuszEQmjYnvseEzJQYE\nyCgWQlyjrLqJfadK2Xe6jKIK3wQyFrOBOeOTmJmZyLjhMRgNg2eSFCGE6O+sIUYWTU5h0eQUqurs\n7D9dxr5TpRw/X8nx85WYjXqy0+OZmZnIhFFxmIxSR4veJS3IYlAyhJjY+OV59p0q5UKxbyQKo0FP\n9ug4ZmYmMnF0HGaTTAk92EnLkZAYGFiKKxtbGjRKq5oAXzI9VUnwNWikxaDXd3kGNYkBIV0s/CRB\nDiKaplFRayevsJa8wjrOFdVSUFqPpvmm5M4cEcvMcYlMyUgg1CI3TYKJnBiFxMDApGkaBaUN7D1V\nwv7TZVTXOwBfl7gxqVGMTolidHIkI4ZGEnKDxg6JAQGSIDeTBHkQc7o8XCypJ6+wlnOFteQV1VHX\n6GzZbtDrUIbHMDk9nmljh/T5LE6i78iJUUgMDHxeTSP3Ug37TpVyLK+yJVkGX32fOiSc9OQoRqdE\nMjolivgoCzrdlVxIYkCA9EEWg4ymaVTW2jlX5Gsdzius5VJZAx7vlQu76HAzU5UERidHkZ4SxfCk\ncJKHRkuFKIQQg4Bep0NJi0FJiwGgqs5OXlGd/65hLfml9eSX1LP1sG//yDAzo5MjSU/xtTRHRFn7\nsPRioJIEWfQ5h9NDea2N8hobFTV232utvWWd0+Vt2deg1zE8KYLRza0FyVHERoa0aS0QQggxeMVG\nWoiNtDB97BAAXG4P+aUNLQlzXlEdR3IrOJJb0fKZ6HAzCdFW4qOsJERbSIi2+pctREeEoJdziLiK\nJMgiYFxuD412N402l+/V7qK+yUV5je1KElxjo94/BvHVrCEGEmNCSYyxMqpV67DJKA/XCSGE8DEZ\nDaSn+M4Rzarq7L4ueIV1lFTbKCpv4FxhLbmXa6/5vNGgIy7KSkKUhfhoXwIdG2Eh3Goi1GIkzGoi\nzGLEGmKURDqISIIcpLyahtfr/+d/7/JouFweXB4vLrfvn9N95b3L47ny3u3F6fK0JL5NzYmww/fa\nZHfjdHs7LINBryMuykJaYkSrisl3RZ8QbSXMYpSWYSGEEF0WG2lhRqSFGeMSW/oguz1equrslNfa\nqaixUV5jp8J/p7K8xt4yasb16MCXMFvaJs7Ny6EhRoxGPSajHrNRj8lowGTwLV/zz6DHbDJg0OvQ\n63XodTr0el93Ejnv9Q+DMkH+m19tw+PpODm7WT3xcGN7X3HNKk3zrdNAQ2vzGc2/QWv1XZp/f82r\n4dXA69XwaJp/WcPr9SXHgdBceYRajCTHh/kqDquJUMuVSiTcaiIh2kJ8lJWYiJAuD9cjhBBCdIfR\noGdITChDYkLb3W5zuKnwJ89V9Q6a7Ffufjba3G2Wq8oduAOUZ+jAlzS3kzjr9Tp0Ot8+zYl08zL4\ntl1Z59uga17RznHaLPfA6bg3knuDXsf9i9LJGhkb0OMMygS5tsGB1xv40Tm6EwfXfKQTQct1/xh8\nG/U6/z460KEHne+PyaBv/oNqtdz8B6ejzbJBr2u5qjUZ9S1Xwc3LZqPhmqtfk0lPmD/5DbWYCA0x\nSsIrhBBiQLKGGBk2JJxhQ8I7tf/Vd1GbHG7cbm+bu7CulmVPO+u8uN1eX2OW/06upml4rmrQan2n\n16v5G+iuajRrbjDzbW9e521pYLtae41xHW7vhN4aFM1g0OFweQJ+nEGZIP/xZ3fICAZCCCGECBiz\nyYDZZCAmIqSviyICQOZuFEIIIYQQohVJkIUQQgghhGhFEmQhhBBCCCFakQRZCCGEEEKIViRBFkII\nIYQQohVJkIUQQgghhGhFEmQhhBBCCCFakQRZCCGEEEKIViRBFkIIIYQQohVJkIUQQgghhGhFEmQh\nhBBCCCFakQRZCCGEEEKIVoy9fUBFUX4NzAK8wHdVVT3YatsS4N8AN7BRVdVnbvQZIYQQQgghelKv\ntiArijIfSFdVdQ7wBPD8Vbs8B9wFzANuUxRlbCc+I4QQQgghRI/p7S4Wi4F1AKqqngGiFUUJB1AU\nZSRQqapqkaqqGrAeWNLRZ4QQQgghhOhpvZ0gJwHlrZYr/Ova21YODAUSO/iMEEIIIYQQPaqvH9LT\ndWNbR58RQgghhBDipvT2Q3pFtG39TQaKW20b2mpbClAIODr4zPXoEhIibq6kYsCTGBAgcSAkBoTE\ngOi63m5B3gTcC6AoyhSgUFXVRgBVVfOBCEVR0hRFMQIr/ftvvt5nhBBCCCGE6Gk6TdN69YCKovwC\nWAB4gO8AU4AaVVU/VBRlHvArQAPeV1X12fY+o6rq8V4ttBBCCCGECBq9niALIYQQQgjRn/X1Q3pC\nCCGEEEL0K5IgCyGEEEII0YokyEIIIYQQQrTS28O8BZSiKL8GZgFe4Luqqh7s4yKJXqIoynh8My7+\nWlXV3yqKkgr8L76LwGLgEVVVXX1ZRhFYiqL8Ct809Qbgl8ABJAaCiqIoVuBNfBNMhQDPAMeQOAg6\niqJYgBPAz4BtSAwEDUVRFgDv4fvvrwNygP+kizEwaFqQFUWZD6SrqjoHeAJ4vo+LJHqJoiih+P57\nb2m1+mfAb1RVXQDkAY/3RdlE71AUZSGQ6f/7vwNYiy8GXpAYCCqrgAOqqi4EHgB+jcRBsPoxUOl/\nL+eD4LNdVdVbVVVdpKrq03QjBgZNggwsxteCiKqqZ4BoRVHC+7ZIopfY8SVFrSeQWQh87H//MbCk\nl8sketcO4D7/+xogDN/QkB/510kMBAFVVf+squp/+RfTgEtIHAQdRVEUYCywHl8L4gLkfBBsrp51\neSFdjIHB1MUiCWjdpaLCv+5c3xRH9BZVVb2Aw1cntghrdfuk7P+3d3chWlRxHMe/kVb0ImZQZoiK\n1F8KupDKxFJCUDPfSEl7I6W6qoveKApCIerOoPAykYxMsqhQyqwolTQqL9SS/oalkpbQjVl4UbZd\nnLMwbgmJrbs98/3AsnNmZofzsH+e5/ecOTPD8U9pVIfJzC7gaG3eS/lgnGoNtFNEfEp5GutM4APr\noHWWUp6zsLC2/Txonysj4m1gCGX0+NyTrYFOGkHuqee3B7WXtdASETGbcursQY7/v1sDLZKZE4BZ\nwKtYB60SEXcDW+rTef+JNdD5vgWWZOYcypek5Rw/IPyvaqCTAvJByohxt2Ecf8pd7XIkIs6uy5dR\n6kMdLCKmAk8C0zLzCNZA60TE2HqBLpm5g3LBpnXQLrcAsyNiK+Vs0tPAr9ZAe2TmwcxcU5e/A34C\nLjzZGuikgLwBmAflTRI4kJm/9W2X1Ic+BObW5bnA+j7si3pZRAyiPKZ+RmYerqutgfaZCDwKEBGX\nAOdT6mBe3W4ddLjMXJCZ4zJzPPAS5fS6NdAiEXFHRHS/Dwyl3NVmBSdZAx31qOmIeI4yGf8Y8EBm\n7uzjLuk0qF+IlgIjgN+BA8CdwMuUWz3tAxZl5rE+66R6VUTcDywGdlNOn3UB91BOrVkDLVFv7bUc\nGA6cAywBtlFu72QdtExELAa+B97HGmiNeoOGVcBgYCDlfWA7sJKTqIGOwIwgOgAAApZJREFUCsiS\nJEnSqeqkKRaSJEnSKTMgS5IkSQ0GZEmSJKnBgCxJkiQ1GJAlSZKkBgOyJEmS1GBAliRJkhoMyJIk\nSVKDAVmSJElqGNDXHZAk/V1EXAusBiZn5t6IGA0sAy4EHs7MrXW/ZcAo4PHM/LrHMR4CZmfmTae3\n95L0/+YIsiT1Q5n5BXAgM/fW9h5gJbCrOxxX24EZPcNx9SXweW/3VZI6jQFZkvqhiBgJ7Omxeh8w\norHP9cBnmdl1gsNMAjb1SgclqYM5xUKS+qeJwMcRMQG4DfgBWAWMBIiIM4GrMnN5bU8BxgCHgemZ\nOR+4kTItg4i4oraHZeYzEbEIOJKZb0TEeGA6ZcT5z8xcGxEXAXOAX4ABmfnaaXrdktTnHEGWpP5p\nIjAc+AZ4BHgeOAhcXLfPB14HiIgRwGOZ+SKwBfgjIgYAF2Tm4br/pcCPwFm1fTOwMyKGASuAZ4H1\nwOC6/a76eweNUWtJagMDsiT1T6OAr4AXMvNY/ekCDkXEOMro75G670JgXV2+AfgEuAbY1n2wzNwI\n3A68UlddnpkJ3EoJ4ROBGUD3SPFbwBTgPWoQl6S2MCBLUj8TEUOBQ5n5DjAyIgZGxLS6eT+wMDPX\nNv5kCOViPSihdjMl8G6uUy+6Dc/M3RExBsiImAwcBdZl5obMfBO4OiKuA+6r0zSeok7rkKS2cA6y\nJPU/Y4GP6vJGYAGwprZ3UecVNywHZkbEEMpc5UmUC/pGA+829tsUEXOBQcDPwHmUEeMnImI6JWhv\npnw27I+IWcCgzFz93748SerfzujqOtHFz5IkSVL7OMVCkiRJajAgS5IkSQ0GZEmSJKnBgCxJkiQ1\nGJAlSZKkBgOyJEmS1GBAliRJkhoMyJIkSVLDX3kVN0nq7Pz0AAAAAElFTkSuQmCC\n",
                        "text/plain": [
                            "<matplotlib.figure.Figure at 0x7f89639cf7b8>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                }
            ],
            "source": [
                "fig = make_distribution_plots(f0, f1)\n",
                "\n",
                "fig.show()"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Losses and costs\n",
                "\n",
                "\n",
                "After observing $z_k, z_{k-1}, \\ldots, z_0$, the decision maker chooses among three distinct actions:\n",
                "\n",
                "* He decides that $x = x_1$ and draws no more $z$'s\n",
                "\n",
                "* He decides that $x = x_0$ and draws no more $z$'s\n",
                "\n",
                "* He postpones deciding now and instead chooses to draw a $z_{k+1}$\n",
                "\n",
                "Associated with these three actions, the decision maker suffers three kinds of losses:\n",
                "\n",
                " \n",
                "* A loss $L_0$ if he decides $x = x_0$ when actually $x=x_1$\n",
                "\n",
                "* A loss $L_1$ if he decides $x = x_1$ when actually $x=x_0$\n",
                "\n",
                "* A cost $c$ if he postpones deciding and chooses instead to draw another  $z$ \n",
                "\n",
                "For example, suppose that we regard $x=x_0$ as a null hypothesis. Then  \n",
                "\n",
                "* We can think of $L_1$ as the loss associated with a type I error\n",
                "\n",
                "* We can think of $L_0$ as the loss associated with a type II error"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### A Bellman equation\n",
                "\n",
                "Let $J_k(p_k)$ be the  total loss for a decision maker who with posterior probability $p_k$ who chooses optimally.\n",
                "\n",
                "The loss functions $\\{J_k(p_k)\\}_k$ satisfy the Bellman equations\n",
                "\n",
                "$$ J_k(p_k) = \\min \\left[ (1-p_k) L_0, p_k L_1, c + E_{z_{k+1}} \\left\\{ J_{k+1} (p_{k+1} \\right\\} \\right] $$\n",
                "\n",
                "where $E_{z_{k+1}}$ denotes a mathematical expectation over the distribution of $z_{k+1}$ and  the minimization is over the three actions, accept $x_0$, accept $x_1$, and postpone deciding and draw \n",
                "a $z_{k+1}$.  \n",
                "\n",
                "Let \n",
                "\n",
                "$$ A_k(p_k) = E_{z_{k+1}} \\left\\{ J_{k+1} \\left[\\frac{ p_k f_0(z_{k+1})}{ p_k f_0(z_{k+1}) + (1-p_k) f_1 (z_{k+1})  } \\right] \\right\\} $$\n",
                "\n",
                "Then we can write out Bellman equation as\n",
                "\n",
                "$$ J_k(p_k) = \\min \\left[ (1-p_k) L_0, p_k L_1, c + A_k(p_k) \\right] $$\n",
                "\n",
                "where $p_k \\in [0,1]$.  \n",
                "\n",
                "Evidently,the optimal decision rule is characterized by two numbers $\\alpha_k, \\beta_k \\in (0,1) \\times (0,1)$\n",
                "that satisfy\n",
                "\n",
                "$$ (1- p_k) L_0 < \\min p_k L_1, c + A_k(p_k)  \\textrm { if } p_k \\geq \\alpha_k  $$\n",
                "\n",
                "and \n",
                "\n",
                "$$ p_k L_1 < \\min (1-p_k) L_0,  c + A_k(p_k) \\textrm { if } p_k \\leq \\beta_k $$\n",
                "\n",
                "The optimal decision rule is then\n",
                "\n",
                "$$ \\textrm { accept } x=x_0 \\textrm{ if } p_k \\geq \\alpha_k \\\\\n",
                "   \\textrm { accept } x=x_1 \\textrm{ if } p_k \\leq \\beta_k \\\\\n",
                "   \\textrm { draw another }  z \\textrm{ if }  \\beta_k \\leq p_k \\leq \\alpha_k $$"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Infinite horizon version\n",
                "\n",
                "An infinite horizon version of this problem is associated with the limiting Bellman equation \n",
                "\n",
                "$$ J(p_k) = \\min \\left[ (1-p_k) L_0, p_k L_1, c + A(p_k) \\right] \\quad (*) $$\n",
                "\n",
                "where\n",
                "\n",
                "$$ A(p_k) = E_{z_{k+1}} \\left\\{ J \\left[\\frac{ p_k f_0(z_{k+1})}{ p_k f_0(z_{k+1}) + (1-p_k) f_1 (z_{k+1})  } \\right] \\right\\} $$\n",
                "\n",
                "and again the minimization is over the three actions, accept $x_1$, accept $x_0$, and postpone deciding and draw \n",
                "a $z_{k+1}$.\n",
                "\n",
                "Here\n",
                "\n",
                "  * $ (1-p_k) L_0$ is the expected loss associated with accepting $x_1$ (i.e., the cost of making a type I error)\n",
                "  \n",
                "  *  $p_k L_1$ is the expected loss associated with accepting $x_0$ (i.e., the cost of making a type II error)\n",
                "  \n",
                "  * $ c + A(p_k)$ is the expected cost associated with drawing one more  $z$\n",
                "\n",
                "\n",
                "Now the optimal decision rule is characterized by two probabilities $0 < \\beta < \\alpha < 1$ and \n",
                "\n",
                "\n",
                "$$ \\textrm { accept } x=x_0 \\textrm{ if } p_k \\geq \\alpha \\\\\n",
                "   \\textrm { accept } x=x_1 \\textrm{ if } p_k \\leq \\beta \\\\\n",
                "   \\textrm { draw another }  z \\textrm{ if } \\beta \\leq p_k \\leq \\alpha $$\n",
                "\n",
                "\n",
                "  "
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "One sensible approach is to write the three components of the value function that appears on the rights side of the Bellman equation as separate functions.  Later, doing this will help us obey the don't repeat yourself (DRY) rule of coding.  Here goes:"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 5,
            "metadata": {
                "collapsed": true
            },
            "outputs": [],
            "source": [
                "def expect_loss_choose_0(p, L0):\n",
                "    \"For a given probability return expected loss of choosing model 0\"\n",
                "    return (1-p)*L0\n",
                "\n",
                "def expect_loss_choose_1(p, L1):\n",
                "    \"For a given probability return expected loss of choosing model 1\"\n",
                "    return p*L1\n",
                "\n",
                "def EJ(p, f0, f1, J):\n",
                "    \"\"\"\n",
                "    We will need to be able to evaluate the expectation of our Bellman\n",
                "    equation J. In order to do this, we need the current probability\n",
                "    that model 0 is correct (p), the distributions (f0, f1), and a\n",
                "    function that can evaluate the Bellman equation\n",
                "    \"\"\"\n",
                "    # Get the current distribution we believe (p*f0 + (1-p)*f1)\n",
                "    curr_dist = p*f0 + (1-p)*f1\n",
                "    \n",
                "    # Get tomorrow's expected distribution through Bayes law\n",
                "    tp1_dist = np.clip((p*f0) / (p*f0 + (1-p)*f1), 0, 1)\n",
                "    \n",
                "    # Evaluate the expectation\n",
                "    EJ = curr_dist @ J(tp1_dist)\n",
                "    \n",
                "    return EJ\n",
                "\n",
                "def expect_loss_cont(p, c, f0, f1, J):\n",
                "    return c + EJ(p, f0, f1, J)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "To approximate the solution of the Bellman equation (\\*) above, we can deploy a method known as value function iteration (iterating on the Bellman equation) on a grid of points. Because we are iterating on a grid, the current probability, $p_k$, is restricted to a set number of points. However, in order to evaluate the expectation of the Bellman equation for tomorrow, $A(p_{k})$, we must be able to evaluate at various $p_{k+1}$ which may or may not correspond with points on our grid. The way that we resolve this issue is by using *linear interpolation*. This means to evaluate $J(p)$ where $p$ is not a grid point, we must use two points: first, we use the largest of all the grid points smaller than $p$, and call it $p_i$, and, second, we use the grid point immediately after $p$, named $p_{i+1}$, to approximate the function value in the following manner:\n",
                "\n",
                "$$ J(p) = J(p_i) + (p - p_i) \\frac{J(p_{i+1}) - J(p_i)}{p_{i+1} - p_{i}}$$\n",
                "\n",
                "In one dimension, you can think of this as simply drawing a line between each pair of points on the grid.\n",
                "\n",
                "For more information on both linear interpolation and value function iteration methods, see the Quant-Econ [lecture](http://quant-econ.net/py/ifp.html) on the income fluctuation problem."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 6,
            "metadata": {
                "collapsed": false
            },
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Iteration    Distance       Elapsed (seconds)\n",
                        "---------------------------------------------\n",
                        "5            8.042e-02      4.618e-02         \n",
                        "10           6.418e-04      7.645e-02         \n",
                        "15           4.482e-06      1.071e-01         \n"
                    ]
                }
            ],
            "source": [
                "def bellman_operator(pgrid, c, f0, f1, L0, L1, J):\n",
                "    \"\"\"\n",
                "    Evaluates the value function for a given continuation value\n",
                "    function; that is, evaluates\n",
                "\n",
                "        J(p) = min(pL0, (1-p)L1, c + E[J(p')])\n",
                "\n",
                "    Uses linear interpolation between points\n",
                "    \"\"\"\n",
                "    m = np.size(pgrid)\n",
                "    assert m == np.size(J)\n",
                "    \n",
                "    J_out = np.zeros(m)\n",
                "    J_interp = interp.UnivariateSpline(pgrid, J, k=1, ext=0)\n",
                "\n",
                "    for (p_ind, p) in enumerate(pgrid):\n",
                "        # Payoff of choosing model 0\n",
                "        p_c_0 = expect_loss_choose_0(p, L0)\n",
                "        p_c_1 = expect_loss_choose_1(p, L1)\n",
                "        p_con = expect_loss_cont(p, c, f0, f1, J_interp)\n",
                "        \n",
                "        J_out[p_ind] = min(p_c_0, p_c_1, p_con)\n",
                "\n",
                "    return J_out\n",
                "\n",
                "# To solve\n",
                "pg = np.linspace(0, 1, 251)\n",
                "bell_op = lambda vf: bellman_operator(pg, 0.5, f0, f1, 5.0, 5.0, vf)\n",
                "J = qe.compute_fixed_point(bell_op, np.zeros(pg.size), error_tol=1e-6,\n",
                "                           verbose=True, print_skip=5)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Now for some gentle criticisms of the preceding code. Although it works fine, by writing the code in terms of  functions, we have to  pass around some things that are constant throughout the problem,  i.e., $c$, $f_0$, $f_1$, $L_0$, and $L_1$. \n",
                "\n",
                "Now that we have a working script, let's turn it into a class.  This will allow us to simplify the function calls and make the code more reusable.\n",
                "\n",
                "So to illustrate a good alternative approach, we write a class that stores all of our parameters for us internally and incorporates  many of the same functions that we used above.  "
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 7,
            "metadata": {
                "collapsed": true
            },
            "outputs": [],
            "source": [
                "# %load -r 22-276 dependencies/Wald_Friedman_utils.py\n",
                "\n",
                "class WaldFriedman(object):\n",
                "    \"\"\"\n",
                "    This class is used to solve the problem presented in the \"Wald Friedman\"\n",
                "    notebook presented on the QuantEcon website.\n",
                "\n",
                "    Parameters\n",
                "    ----------\n",
                "    c : scalar(Float64)\n",
                "        Cost of postponing decision\n",
                "    L0 : scalar(Float64)\n",
                "        Cost of choosing model 0 when the truth is model 1\n",
                "    L1 : scalar(Float64)\n",
                "        Cost of choosing model 1 when the truth is model 0\n",
                "    f0 : array_like(Float64)\n",
                "        A finite state probability distribution\n",
                "    f1 : array_like(Float64)\n",
                "        A finite state probability distribution\n",
                "    m : scalar(Int)\n",
                "        Number of points to use in function approximation\n",
                "    \"\"\"\n",
                "    def __init__(self, c, L0, L1, f0, f1, m=25):\n",
                "        self.c = c\n",
                "        self.L0, self.L1 = L0, L1\n",
                "        self.m = m\n",
                "        self.pgrid = np.linspace(0.0, 1.0, m)\n",
                "\n",
                "        # Renormalize distributions so nothing is \"too\" small\n",
                "        f0 = np.clip(f0, 1e-8, 1-1e-8)\n",
                "        f1 = np.clip(f1, 1e-8, 1-1e-8)\n",
                "        self.f0 = f0 / np.sum(f0)\n",
                "        self.f1 = f1 / np.sum(f1)\n",
                "        self.J = np.zeros(m)\n",
                "\n",
                "    def current_distribution(self, p):\n",
                "        \"\"\"\n",
                "        This function takes a value for the probability with which\n",
                "        the correct model is model 0 and returns the mixed\n",
                "        distribution that corresponds with that belief.\n",
                "        \"\"\"\n",
                "        return p*self.f0 + (1-p)*self.f1\n",
                "\n",
                "    def bayes_update_k(self, p, k):\n",
                "        \"\"\"\n",
                "        This function takes a value for p, and a realization of the\n",
                "        random variable and calculates the value for p tomorrow.\n",
                "        \"\"\"\n",
                "        f0_k = self.f0[k]\n",
                "        f1_k = self.f1[k]\n",
                "\n",
                "        p_tp1 = p*f0_k / (p*f0_k + (1-p)*f1_k)\n",
                "\n",
                "        return np.clip(p_tp1, 0, 1)\n",
                "\n",
                "    def bayes_update_all(self, p):\n",
                "        \"\"\"\n",
                "        This is similar to `bayes_update_k` except it returns a\n",
                "        new value for p for each realization of the random variable\n",
                "        \"\"\"\n",
                "        return np.clip(p*self.f0 / (p*self.f0 + (1-p)*self.f1), 0, 1)\n",
                "\n",
                "    def payoff_choose_f0(self, p):\n",
                "        \"For a given probability specify the cost of accepting model 0\"\n",
                "        return (1-p)*self.L0\n",
                "\n",
                "    def payoff_choose_f1(self, p):\n",
                "        \"For a given probability specify the cost of accepting model 1\"\n",
                "        return p*self.L1\n",
                "\n",
                "    def EJ(self, p, J):\n",
                "        \"\"\"\n",
                "        This function evaluates the expectation of the value function\n",
                "        at period t+1. It does so by taking the current probability\n",
                "        distribution over outcomes:\n",
                "\n",
                "            p(z_{k+1}) = p_k f_0(z_{k+1}) + (1-p_k) f_1(z_{k+1})\n",
                "\n",
                "        and evaluating the value function at the possible states\n",
                "        tomorrow J(p_{t+1}) where\n",
                "\n",
                "            p_{t+1} = p f0 / ( p f0 + (1-p) f1)\n",
                "\n",
                "        Parameters\n",
                "        ----------\n",
                "        p : Scalar(Float64)\n",
                "            The current believed probability that model 0 is the true\n",
                "            model.\n",
                "        J : Function\n",
                "            The current value function for a decision to continue\n",
                "\n",
                "        Returns\n",
                "        -------\n",
                "        EJ : Scalar(Float64)\n",
                "            The expected value of the value function tomorrow\n",
                "        \"\"\"\n",
                "        # Pull out information\n",
                "        f0, f1 = self.f0, self.f1\n",
                "\n",
                "        # Get the current believed distribution and tomorrows possible dists\n",
                "        # Need to clip to make sure things don't blow up (go to infinity)\n",
                "        curr_dist = self.current_distribution(p)\n",
                "        tp1_dist = self.bayes_update_all(p)\n",
                "\n",
                "        # Evaluate the expectation\n",
                "        EJ = curr_dist @ J(tp1_dist)\n",
                "\n",
                "        return EJ\n",
                "\n",
                "    def payoff_continue(self, p, J):\n",
                "        \"\"\"\n",
                "        For a given probability distribution and value function give\n",
                "        cost of continuing the search for correct model\n",
                "        \"\"\"\n",
                "        return self.c + self.EJ(p, J)\n",
                "\n",
                "    def bellman_operator(self, J):\n",
                "        \"\"\"\n",
                "        Evaluates the value function for a given continuation value\n",
                "        function; that is, evaluates\n",
                "\n",
                "            J(p) = min( (1-p)L0, pL1, c + E[J(p')])\n",
                "\n",
                "        Uses linear interpolation between points\n",
                "        \"\"\"\n",
                "        payoff_choose_f0 = self.payoff_choose_f0\n",
                "        payoff_choose_f1 = self.payoff_choose_f1\n",
                "        payoff_continue = self.payoff_continue\n",
                "        c, L0, L1, f0, f1 = self.c, self.L0, self.L1, self.f0, self.f1\n",
                "        m, pgrid = self.m, self.pgrid\n",
                "\n",
                "        J_out = np.empty(m)\n",
                "        J_interp = interp.UnivariateSpline(pgrid, J, k=1, ext=0)\n",
                "\n",
                "        for (p_ind, p) in enumerate(pgrid):\n",
                "            # Payoff of choosing model 0\n",
                "            p_c_0 = payoff_choose_f0(p)\n",
                "            p_c_1 = payoff_choose_f1(p)\n",
                "            p_con = payoff_continue(p, J_interp)\n",
                "\n",
                "            J_out[p_ind] = min(p_c_0, p_c_1, p_con)\n",
                "\n",
                "        return J_out\n",
                "\n",
                "    def solve_model(self, tol=1e-7):\n",
                "        J =  qe.compute_fixed_point(self.bellman_operator, np.zeros(self.m),\n",
                "                                    error_tol=tol, verbose=False)\n",
                "\n",
                "        self.J = J\n",
                "        return J\n",
                "\n",
                "    def find_cutoff_rule(self, J):\n",
                "        \"\"\"\n",
                "        This function takes a value function and returns the corresponding\n",
                "        cutoffs of where you transition between continue and choosing a\n",
                "        specific model\n",
                "        \"\"\"\n",
                "        payoff_choose_f0 = self.payoff_choose_f0\n",
                "        payoff_choose_f1 = self.payoff_choose_f1\n",
                "        m, pgrid = self.m, self.pgrid\n",
                "\n",
                "        # Evaluate cost at all points on grid for choosing a model\n",
                "        p_c_0 = payoff_choose_f0(pgrid)\n",
                "        p_c_1 = payoff_choose_f1(pgrid)\n",
                "\n",
                "        # The cutoff points can be found by differencing these costs with\n",
                "        # the Bellman equation (J is always less than or equal to p_c_i)\n",
                "        lb = pgrid[np.searchsorted(p_c_1 - J, 1e-10) - 1]\n",
                "        ub = pgrid[np.searchsorted(J - p_c_0, -1e-10)]\n",
                "\n",
                "        return (lb, ub)\n",
                "\n",
                "    def simulate(self, f, p0=0.5):\n",
                "        \"\"\"\n",
                "        This function takes an initial condition and simulates until it\n",
                "        stops (when a decision is made).\n",
                "        \"\"\"\n",
                "        # Check whether vf is computed\n",
                "        if np.sum(self.J) < 1e-8:\n",
                "            self.solve_model()\n",
                "\n",
                "        # Unpack useful info\n",
                "        lb, ub = self.find_cutoff_rule(self.J)\n",
                "        update_p = self.bayes_update_k\n",
                "        curr_dist = self.current_distribution\n",
                "        drv = qe.discrete_rv.DiscreteRV(f)\n",
                "\n",
                "        # Initialize a couple useful variables\n",
                "        decision_made = False\n",
                "        p = p0\n",
                "        t = 0\n",
                "\n",
                "        while decision_made is False:\n",
                "            # Maybe should specify which distribution is correct one so that\n",
                "            # the draws come from the \"right\" distribution\n",
                "            k = drv.draw()[0]\n",
                "            t = t+1\n",
                "            p = update_p(p, k)\n",
                "            if p < lb:\n",
                "                decision_made = True\n",
                "                decision = 1\n",
                "            elif p > ub:\n",
                "                decision_made = True\n",
                "                decision = 0\n",
                "\n",
                "        return decision, p, t\n",
                "\n",
                "    def simulate_tdgp_f0(self, p0=0.5):\n",
                "        \"\"\"\n",
                "        Uses the distribution f0 as the true data generating\n",
                "        process\n",
                "        \"\"\"\n",
                "        decision, p, t = self.simulate(self.f0, p0)\n",
                "\n",
                "        if decision == 0:\n",
                "            correct = True\n",
                "        else:\n",
                "            correct = False\n",
                "\n",
                "        return correct, p, t\n",
                "\n",
                "    def simulate_tdgp_f1(self, p0=0.5):\n",
                "        \"\"\"\n",
                "        Uses the distribution f1 as the true data generating\n",
                "        process\n",
                "        \"\"\"\n",
                "        decision, p, t = self.simulate(self.f1, p0)\n",
                "\n",
                "        if decision == 1:\n",
                "            correct = True\n",
                "        else:\n",
                "            correct = False\n",
                "\n",
                "        return correct, p, t\n",
                "\n",
                "    def stopping_dist(self, ndraws=250, tdgp=\"f0\"):\n",
                "        \"\"\"\n",
                "        Simulates repeatedly to get distributions of time needed to make a\n",
                "        decision and how often they are correct.\n",
                "        \"\"\"\n",
                "        if tdgp==\"f0\":\n",
                "            simfunc = self.simulate_tdgp_f0\n",
                "        else:\n",
                "            simfunc = self.simulate_tdgp_f1\n",
                "\n",
                "        # Allocate space\n",
                "        tdist = np.empty(ndraws, int)\n",
                "        cdist = np.empty(ndraws, bool)\n",
                "\n",
                "        for i in range(ndraws):\n",
                "            correct, p, t = simfunc()\n",
                "            tdist[i] = t\n",
                "            cdist[i] = correct\n",
                "\n",
                "        return cdist, tdist\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Now let's use our class  solve the Bellman equation (*) and check whether it gives same answer attained above."
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 8,
            "metadata": {
                "collapsed": false
            },
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Iteration    Distance       Elapsed (seconds)\n",
                        "---------------------------------------------\n",
                        "5            8.042e-02      4.080e-02         \n",
                        "10           6.418e-04      7.262e-02         \n",
                        "15           4.482e-06      1.044e-01         \n",
                        "\n",
                        "If this is true then both approaches gave same answer\n",
                        "True\n"
                    ]
                }
            ],
            "source": [
                "wf = WaldFriedman(0.5, 5.0, 5.0, f0, f1, m=251)\n",
                "wfJ = qe.compute_fixed_point(wf.bellman_operator, np.zeros(251),\n",
                "                             error_tol=1e-6, verbose=True, print_skip=5)\n",
                "\n",
                "print(\"\\nIf this is true then both approaches gave same answer\")\n",
                "print(np.allclose(J, wfJ))"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "### Numerical Example\n",
                "\n",
                "Now let's specify the two probability distibutions (the ones that we plotted earlier)\n",
                "\n",
                "* for $f_0$ we'll assume a beta distribution with parameters $a=1, b=1$\n",
                "\n",
                "* for $f_1$ we'll assume a beta distribution with parameters $a=9, b=9$\n",
                "\n",
                "The density of a  beta probability distribution with parameters $a$ and $b$ is\n",
                "\n",
                "$$ f(z; a, b) = \\frac{\\Gamma(a+b) z^{a-1} (1-z)^{b-1}}{\\Gamma(a) \\Gamma(b)}$$\n",
                "\n",
                "where $\\Gamma$ is the gamma function \n",
                "\n",
                "$$\\Gamma(t) = \\int_{0}^{\\infty} x^{t-1} e^{-x} dx$$\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 9,
            "metadata": {
                "collapsed": false
            },
            "outputs": [],
            "source": [
                "# Choose parameters\n",
                "c = 1.25\n",
                "L0 = 27.0\n",
                "L1 = 27.0\n",
                "\n",
                "# Choose n points and distributions\n",
                "m = 251\n",
                "# f0 = np.ones(n)/n\n",
                "f0 = st.beta.pdf(np.linspace(0, 1, m), a=2.5, b=3)\n",
                "f0 = f0 / np.sum(f0)\n",
                "f1 = st.beta.pdf(np.linspace(0, 1, m), a=3, b=2.5)\n",
                "f1 = f1 / np.sum(f1)  # Make sure sums to 1"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 10,
            "metadata": {
                "collapsed": false
            },
            "outputs": [],
            "source": [
                "# Create an instance of our WaldFriedman class\n",
                "wf = WaldFriedman(c, L0, L1, f0, f1, m=m)"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 11,
            "metadata": {
                "collapsed": false
            },
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "Iteration    Distance       Elapsed (seconds)\n",
                        "---------------------------------------------\n",
                        "5            9.200e-01      5.442e-02         \n",
                        "10           4.523e-01      1.026e-01         \n",
                        "15           1.223e-01      1.468e-01         \n",
                        "20           2.126e-02      1.917e-01         \n",
                        "25           3.316e-03      2.690e-01         \n",
                        "30           4.752e-04      3.602e-01         \n",
                        "35           6.826e-05      4.373e-01         \n",
                        "40           9.806e-06      4.816e-01         \n",
                        "45           1.409e-06      5.392e-01         \n"
                    ]
                }
            ],
            "source": [
                "# Solve using qe's `compute_fixed_point` function\n",
                "J = qe.compute_fixed_point(wf.bellman_operator, np.zeros(m),\n",
                "                           error_tol=1e-6, verbose=True, print_skip=5)"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "The value function equals $ p L_1$ for $p \\leq \\alpha$, and $(1-p )L_0$ for $ p \\geq \\beta$.\n",
                "Thus, the slopes of the two linear pieces of the value function are determined by $L_1$ and \n",
                "$- L_0$.  \n",
                "\n",
                "The value function is smooth in the interior region in which the probability assigned to distribution  $f_0$ is in the indecisive region $p \\in (\\alpha, \\beta)$.\n",
                "\n",
                "The decision maker continues to sample until the probability that he attaches to model $f_0$ falls below $\\alpha$ or above $\\beta$.\n",
                "\n",
                "The value function is smooth in the interior region in which the probability assigned to distribution  $f_0$ is in the indecisive region $p \\in (\\alpha, \\beta)$.\n",
                "\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "Now to have some fun, you can use the slider and watch the effects on the smoothness of the\n",
                "of the value function in the  middle range as you increase the numbers of functions in the \n",
                "piecewise linear approximation.  \n",
                "\n",
                "The slider let's you choose the cost parameters $L_0, L_1, c$, the parameters of two beta distributions $f_0$ and $f_1$, and the number of points and linear functions $m$ to use in our piece-wise continuous approximation to the value function.\n",
                "\n",
                "It then draws a number of simulations from $f_0$, computes a distribution of waiting times to making a decision, and displays a histogram of correct and incorrect decisions. (Here the correct decision occurs when $p_k$ eventually exceeds $\\beta$.\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 12,
            "metadata": {
                "collapsed": false,
                "scrolled": true
            },
            "outputs": [
                {
                    "data": {
                        "image/png": "iVBORw0KGgoAAAANSUhEUgAABigAAAPnCAYAAAC1HcTSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAANDd/wcJ5PI8aVijfQAAAABJRU5ErkJggg==\n",
                        "text/plain": [
                            "<matplotlib.figure.Figure at 0x7f895f81a400>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                }
            ],
            "source": [
                "col_slide = list(map(convert_rgb_hex, sb.color_palette(\"dark\", 7)))\n",
                "col_num = list(map(convert_rgb_hex, sb.color_palette(\"hls\", 7)))\n",
                "sliders = map(lambda a,b,c,d,e,f,g: widgets.FloatSlider(min=a, max=b,\n",
                "                                                        step=c, value=d,\n",
                "                                                        slider_color=e, \n",
                "                                                        color=f,\n",
                "                                                        description=g),\n",
                "             [0.5, 5.0, 5.0, 1.0, 1.0, 1.0, 1.0],\n",
                "             [2.5, 50.0,\n",
                "              50.0, 9.0, 9.0, 9.0, 9.0],\n",
                "             [0.25, 2.5, 2.5, 0.5, 0.5, 0.5, 0.5],\n",
                "             [1.25, 27.5, 27.5, 2.0, 2.5, 2.5, 2.0],\n",
                "             col_num, col_num,\n",
                "             [\"c\", \"L0\", \"L1\", \"a0\", \"b0\", \"a1\", \"b1\"])\n",
                "cslide, L0slide, L1slide, a0slide, b0slide, a1slide, b1slide = list(sliders)\n",
                "mslide = widgets.IntSlider(min=15, max=251, step=2, \n",
                "                           value=133, description=\"m\")\n",
                "\n",
                "interact(all_param_interact, c=cslide, L0=L0slide, L1=L1slide, a0=a0slide,\n",
                "         b0=b0slide, a1=a1slide, b1=b1slide, m=mslide)\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {
                "collapsed": true
            },
            "source": [
                "### Comparison with Neyman-Pearson formulation\n",
                "\n",
                "For several reasons, it is useful to describe the theory underlying test that Navy Captain G. S. Schuyler had been told to use and that led him to approach Milton Friedman and Allan Wallis to convey his conjecture that superior practical procedures existed.  Evidently,\n",
                "the Navy had told Captail Schuyler to use what it knew to be  a state-of-the-art Neyman-Pearson test. \n",
                "\n",
                "We'll rely on Abraham Wald's elegant summary of Neyman-Pearson theory.  For our purposes, watch for  there features of the setup:\n",
                "\n",
                "   * the assumption of a *fixed *sample size $n$\n",
                "   \n",
                "   * the application of  laws of large numbers, conditioned on alternative probability models, to interpret the probabilities $\\alpha$ and $\\beta$ defined in the Neyman-Pearson theory.\n",
                "   \n",
                "Recall that in the sequential analytic formulation above, that\n",
                "\n",
                "  * The sample size $n$ is not fixed but rather an object to be chosen; technically $n$ is a random variable. \n",
                "  \n",
                "  * The parameters $\\alpha$ and $\\beta$ characterize cut-off rules used to determine $n$ as a random variable.\n",
                "  \n",
                "  * Laws of large numbers make no appearances in the sequential construction.\n",
                "\n",
                "In chapter 1 of **Sequential Analysis** Abraham Wald summarizes the Neyman-Pearson approach to hypothesis testing.\n",
                "\n",
                "Wald frames the problem as making a decision about a probability distribution that is partially known. (You have to assume that *something* is already known in order to state a well posed problem.  Usually, *something* means *a lot*.)\n",
                "\n",
                "By restricting what is unknown, Wald uses the following simple structure to illustrate the main ideas.\n",
                "\n",
                "   * a decision maker wants to decide which of two distributions $f_0$,  $f_1$ govern an i.i.d. random variable $z$\n",
                "   \n",
                "   * The null hypothesis $H_0$ is the statement that $f_0$ governs the data.\n",
                "   \n",
                "   * The alternative hypothesis $H_1$ is the statement that $f_1$ governs the data. \n",
                "   \n",
                "   * The problem is to devise and analyze a test of hypthothesis $H_0$ against the alternative hypothesis $H_1$ on the basis of a sample of a fixed number $n$ independent observations $z_1, z_2, \\ldots, z_n$ of the random variable $z$.  \n",
                "   \n",
                "To quote Abraham Wald,\n",
                "\n",
                "  * *A test procedure leading to the acceptance or rejection of the hypothesis in question is simply a rule specifying, for each possible sample of size $n$, whether the hypothesis should be accepted or rejected on the basis of the sample. This may also be expressed as follows: A test procedure is simply a subdivision of the totality of all possibsle samples of size $n$ into two mutually exclusive parts, say part 1 and part 2, together with the application of the rule that the hypothesis be accepted if the observed sample is contained in part 2.  Part 1 is also called the critical region.  Since part 2 is the totality of all samples of size 2 which are not included in part 1, part 2 is uniquely determined by part 1.  Thus, choosing a test procedure is equivalent to determining a critical region.* \n",
                "  \n",
                "Let's listen to Wald longer:\n",
                "\n",
                "  * *As a basis for choosing among critical regions the following considerations have been advanced by Neyman and Pearson: In accepting or rejecting $H_0$ we may commit errors of two kinds. We commit an error of the first kind if we reject $H_0$ when it is true; we commit an error of the second kind if we accept $H_0$ whe $H_1$ is true.  After a particular critical region $W$ has been chosen, the probability of committing an error of the first kind, as well as the probability of committing an error of the second kind is uniquely determined.  The probability of committing an error of the first kind is equal to the probability, determined by the assumption that $H_0$ is true, that the observed sample will be included in the critical region $W$.  The probability of committing an error of the second kind is equal to the probability, determined on the assumption that $H_1$ is true, that the probability will fall outside the critical region $W$.  For any given critical region $W$ we shall denote the probability of an error of the first kind by $\\alpha$ and the probability of an error of the second kind by $\\beta$.*\n",
                "\n",
                "\n",
                "Let's listen carefully to how Wald applies a law of large numbers to interpret $\\alpha$ and $\\beta$:\n",
                "\n",
                "  * *The probabilities $\\alpha$ and $\\beta$ have the following important practical interpretation: Suppose that we draw a large number of samples of size $n$.  Let $M$ be the number of such samples drawn.  Suppose that for each of these $M$ samples we reject $H_0$ if the sample is included in $W$ and accept $H_0$ if the sample lies outside $W$.  In this way we make $M$ statements of rejection or acceptance.  Some of these statements will in general be wrong.  If $H_0$ is true and if $M$ is large, the probability is nearly $1$ (i.e., it is practically certain) that the proportion of wrong statements (i.e., the number of wrong statements divided by $M$) will be approximately $\\alpha$.  If $H_1$ is true, the probability is nearly $1$ that the proportion of wrong statements will be approximately $\\beta$.  Thus, we can say that in the long run [ here Wald applies a law of large numbers by driving $M \\rightarrow \\infty$ (our comment, not Wald's) ] the proportion of wrong statements will be $\\alpha$ if $H_0$is true and $\\beta $ if $H_1$ is true.*\n",
                "\n",
                "The quantity $\\alpha$ is called the *size* of the critical region, and the quantity $1-\\beta$ is called the *power* of the critical region.\n",
                "\n",
                "Wald notes that \n",
                "\n",
                "  * *one critical region $W$ is more desirable than another if it has smaller values of $\\alpha$ and $\\beta$.  Although either $\\alpha$ or $\\beta$ can be made arbitrarily small by a proper choice of the critical region $W$, it is possible to make both $\\alpha$ and $\\beta$ arbitrarily small for a fixed value of $n$, i.e., a fixed sample size.*\n",
                "\n",
                "\n",
                "Wald summarizes Neyman and Pearson's setup as follows:\n",
                "\n",
                " * *Neyman and Pearson show that a region consisting of all samples $(z_1, z_2, \\ldots, z_n)$ which satisfy the inequality \n",
                " $$\\frac{ f_1(z_1) \\cdots f_1(z_n)}{f_0(z_1) \\cdots f_1(z_n)} \\geq k $$ is a most powerful critical region for testing the hypothesis $H_0$ against the alternative hypothesis $H_1$. The term $k$ on the right side is a constant chosen so that the region will have the required size $\\alpha$.*\n",
                " \n",
                "Wald goes on to discuss Neyman and Pearson's concept of *uniformly most powerful* test. \n",
                "\n",
                "\n",
                "Here is how Wald introduces the notion of a sequential test\n",
                "•A rule is given for making one of the following three decisions at any stage of the experiment (at the m th trial for each integral value of m ): (1) to accept the hypothesis H , (2) to reject the hypothesis H , (3) to continue the experiment by making an additional observation. Thus, such a test procedure is carried out sequentially. On the basis of the first observation one of the aforementioned decisions is made. If the first or second decision is made, the process is terminated. If the third decision is made, a second trial is performed. Again, on the basis of the first two observations one of the three decisions is made. If the third decision is made, a third trial is performed, and so on. The process is continued until either the first or the second decisions is made. The number n  of observations required by such a test procedure is a random variable, since the value of n  depends on the outcome of the observations.\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": null,
            "metadata": {
                "collapsed": true
            },
            "outputs": [],
            "source": []
        }
    ],
    "metadata": {
        "anaconda-cloud": {},
        "kernelspec": {
            "display_name": "Python [Root]",
            "language": "python",
            "name": "Python [Root]"
        },
        "language_info": {
            "codemirror_mode": {
                "name": "ipython",
                "version": 3
            },
            "file_extension": ".py",
            "mimetype": "text/x-python",
            "name": "python",
            "nbconvert_exporter": "python",
            "pygments_lexer": "ipython3",
            "version": "3.5.2"
        }
    },
    "nbformat": 4,
    "nbformat_minor": 0
}