forked from ryanmorrison/bayesian_network
-
Notifications
You must be signed in to change notification settings - Fork 0
/
site5_output_processing.R
430 lines (375 loc) · 22 KB
/
site5_output_processing.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
#### This script is used to compare model outputs of each scenario to existing conditions. Two approaches are used: ####
# 1) Not including cells that are no longer inundated into the analyses
# 2) Assigning cells that are no longer inundated a zero percent recruitment potential and including the cells in the analyses
#### Approach #1 ####
#### Set working directory ####
setwd("/Users/Morrison/Dropbox/Gila Bayesian/bayesian_network")
# Load libraries
library(sp) # vector data
library(raster) # raster data
library(rgdal) # input/output, projections
library(graphics)
library(plyr)
#### Load model results ####
E_input <- read.table("data/site5_1000_to_4000.txt", header=FALSE)
E_output <- read.csv("output/site5/E_q_all_prob_mn.csv", header=TRUE)
CUFA_150_output <- read.csv("output/site5/CUFA_150_q_all_prob_mn.csv", header=TRUE)
CUFA_nomin_output <- read.csv("output/site5/CUFA_nomin_q_all_prob_mn.csv", header=TRUE)
nrow(E_output)
nrow(CUFA_150_output)
nrow(CUFA_nomin_output)
#### Match the number of cells between existing conditions and each scenario ####
E_output_1 <- E_output[E_output[,1]%in%CUFA_150_output[,1],]
E_output_2 <- E_output[E_output[,1]%in%CUFA_nomin_output[,1],]
#### Calculate the mean probability difference between existing conditons and each scenario ####
# Disregarding cells that are no longer inundated
CUFA_150_diff <- (CUFA_150_output - E_output_1)*100
mean(CUFA_150_diff[,2], na.rm=TRUE)
CUFA_nomin_diff <- (CUFA_nomin_output - E_output_2)*100
mean(CUFA_nomin_diff[,2], na.rm=TRUE)
#### Quick histogram plots of the probability differences ####
hist(CUFA_150_diff[,2])
hist(CUFA_nomin_diff[,2])
#### Load all the probabilities for each Q-bin ####
load("output/site5/E_q1_cell_probs.Rdata")
load("output/site5/E_q2_cell_probs.Rdata")
load("output/site5/E_q3_cell_probs.Rdata")
load("output/site5/E_q4_cell_probs.Rdata")
load("output/site5/E_q5_cell_probs.Rdata")
load("output/site5/E_q6_cell_probs.Rdata")
load("output/site5/E_q7_cell_probs.Rdata")
load("output/site5/CUFA_150_q1_cell_probs.Rdata")
load("output/site5/CUFA_150_q2_cell_probs.Rdata")
load("output/site5/CUFA_150_q3_cell_probs.Rdata")
load("output/site5/CUFA_150_q4_cell_probs.Rdata")
load("output/site5/CUFA_150_q5_cell_probs.Rdata")
load("output/site5/CUFA_150_q6_cell_probs.Rdata")
load("output/site5/CUFA_150_q7_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q1_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q2_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q3_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q4_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q5_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q6_cell_probs.Rdata")
load("output/site5/CUFA_nomin_q7_cell_probs.Rdata")
load("output/site5/E_evidence1.Rdata")
load("output/site5/E_evidence2.Rdata")
load("output/site5/E_evidence3.Rdata")
load("output/site5/E_evidence4.Rdata")
load("output/site5/E_evidence5.Rdata")
load("output/site5/E_evidence6.Rdata")
load("output/site5/E_evidence7.Rdata")
load("output/site5/CUFA_150_evidence1.Rdata")
load("output/site5/CUFA_150_evidence2.Rdata")
load("output/site5/CUFA_150_evidence3.Rdata")
load("output/site5/CUFA_150_evidence4.Rdata")
load("output/site5/CUFA_150_evidence5.Rdata")
load("output/site5/CUFA_150_evidence6.Rdata")
load("output/site5/CUFA_150_evidence7.Rdata")
load("output/site5/CUFA_nomin_evidence1.Rdata")
load("output/site5/CUFA_nomin_evidence2.Rdata")
load("output/site5/CUFA_nomin_evidence3.Rdata")
load("output/site5/CUFA_nomin_evidence4.Rdata")
load("output/site5/CUFA_nomin_evidence5.Rdata")
load("output/site5/CUFA_nomin_evidence6.Rdata")
load("output/site5/CUFA_nomin_evidence7.Rdata")
#### Calculate the number of events (length) of each cell ####
# Existing conditions
E_all_probs <- c(E_q1_cell_probs, E_q2_cell_probs, E_q3_cell_probs, E_q4_cell_probs, E_q5_cell_probs, E_q6_cell_probs, E_q7_cell_probs) # Combine probabilities for all cells
head(E_all_probs) # Examine cell probs
E_lengths <- as.data.frame(sapply(E_all_probs, length)) # Calculate the number of events in each cell and convert to data frame
E_lengths <- data.frame(c(as.numeric(row.names(E_lengths))), E_lengths[,1]) # Create new data frame with cell number in first column and number of events in second colume
colnames(E_lengths) <- c("cell", "length") # Name columns
head(E_lengths) # Preview the data table
nrow(E_lengths) # Count number of cells (rows)
E_mean <- mean(E_lengths[,2]) # Calculate the mean number of events for all cells
E_mean # View mean
# CUFA 150
CUFA_150_all_probs <- c(CUFA_150_q1_cell_probs, CUFA_150_q2_cell_probs, CUFA_150_q3_cell_probs, CUFA_150_q4_cell_probs, CUFA_150_q5_cell_probs, CUFA_150_q6_cell_probs, CUFA_150_q7_cell_probs) # Combine probabilities for all cells
head(CUFA_150_all_probs) # Examine cell probs
CUFA_150_lengths <- as.data.frame(sapply(CUFA_150_all_probs, length)) # Calculate the number of events in each cell and convert to data frame
CUFA_150_lengths <- data.frame(c(as.numeric(row.names(CUFA_150_lengths))), CUFA_150_lengths[,1]) # Create new data frame with cell number in first column and number of events in second colume
colnames(CUFA_150_lengths) <- c("cell", "length") # Name columns
head(CUFA_150_lengths) # Preview the data table
nrow(CUFA_150_lengths) # Count number of cells (rows)
CUFA_150_mean <- mean(CUFA_150_lengths[,2]) # Calculate the mean number of events for all cells
CUFA_150_mean # View mean
# CUFA no minimum
CUFA_nomin_all_probs <- c(CUFA_nomin_q1_cell_probs, CUFA_nomin_q2_cell_probs, CUFA_nomin_q3_cell_probs, CUFA_nomin_q4_cell_probs, CUFA_nomin_q5_cell_probs, CUFA_nomin_q6_cell_probs, CUFA_nomin_q7_cell_probs) # Combine probabilities for all cells
head(CUFA_nomin_all_probs) # Examine cell probs
CUFA_nomin_lengths <- as.data.frame(sapply(CUFA_nomin_all_probs, length)) # Calculate the number of events in each cell and convert to data frame
CUFA_nomin_lengths <- data.frame(c(as.numeric(row.names(CUFA_nomin_lengths))), CUFA_nomin_lengths[,1]) # Create new data frame with cell number in first column and number of events in second colume
colnames(CUFA_nomin_lengths) <- c("cell", "length") # Name columns
head(CUFA_nomin_lengths) # Preview the data table
nrow(CUFA_nomin_lengths) # Count number of cells (rows)
CUFA_nomin_mean <- mean(CUFA_nomin_lengths[,2]) # Calculate the mean number of events for all cells
CUFA_nomin_mean # View mean
#### Calculate the difference in events between existing conditions and the scenario for each cell ####
E_lengths_1 <- E_lengths[E_lengths[ ,1] %in% CUFA_150_lengths[ ,1], ] # Match cells in existing conditions and CUFA 150
E_lengths_2 <- E_lengths[E_lengths[ ,1] %in% CUFA_nomin_lengths[ ,1], ] # Match cells in existing conditions and CUFA no mininum
# CUFA 150
CUFA_150_pct_diff <- as.data.frame((E_lengths_1[,2]-CUFA_150_lengths[,2])/E_lengths_1[,2]*100)
CUFA_150_pct_diff <- data.frame((CUFA_150_lengths[,1]), CUFA_150_pct_diff[,1])
colnames(CUFA_150_pct_diff) <- c("cell","pct_decrease")
CUFA_150_pct_diff <- CUFA_150_pct_diff[order(CUFA_150_pct_diff$cell),]
head(CUFA_150_pct_diff)
mean(CUFA_150_pct_diff[,2])
# CUFA no minimum
CUFA_nomin_pct_diff <- as.data.frame((E_lengths_2[,2]-CUFA_nomin_lengths[,2])/E_lengths_2[,2]*100)
CUFA_nomin_pct_diff <- data.frame((CUFA_nomin_lengths[,1]), CUFA_nomin_pct_diff[,1])
colnames(CUFA_nomin_pct_diff) <- c("cell","pct_decrease")
CUFA_nomin_pct_diff <- CUFA_nomin_pct_diff[order(CUFA_nomin_pct_diff$cell),]
head(CUFA_nomin_pct_diff)
mean(CUFA_nomin_pct_diff[,2])
#### Define a utility function that accounts for a decrease in events for a particular cell ####
utility_ftn <- function(X) (100-X)/100
#### Apply the utility function to each scenario ####
# CUFA 150
CUFA_150_utility <- utility_ftn(CUFA_150_pct_diff[,2])
CUFA_150_utility <- cbind(CUFA_150_pct_diff[,1], CUFA_150_utility)
colnames(CUFA_150_utility) <- c("cell","pct_of_network_output")
head(CUFA_150_utility)
CUFA_nomin_utility <- utility_ftn(CUFA_nomin_pct_diff[,2])
CUFA_nomin_utility <- cbind(CUFA_nomin_pct_diff[,1], CUFA_nomin_utility)
colnames(CUFA_nomin_utility) <- c("cell","pct_of_network_output")
head(CUFA_nomin_utility)
CUFA_150_utility_probs <- CUFA_150_output[,2] * CUFA_150_utility[,2]
CUFA_150_utility_probs <- cbind(CUFA_150_pct_diff[,1], CUFA_150_utility_probs)
colnames(CUFA_150_utility_probs) <- c("cell","mean_prob")
head(CUFA_150_utility_probs)
head(CUFA_150_output)
mean(CUFA_150_utility_probs[,2], na.rm=TRUE)
mean(E_output[,2])
mean(CUFA_150_utility_probs[,2], na.rm=TRUE) - mean(E_output[,2])
CUFA_nomin_utility_probs <- CUFA_nomin_output[,2] * CUFA_nomin_utility[,2]
CUFA_nomin_utility_probs <- cbind(CUFA_nomin_pct_diff[,1], CUFA_nomin_utility_probs)
colnames(CUFA_nomin_utility_probs) <- c("cell","mean_prob")
head(CUFA_nomin_utility_probs)
head(CUFA_nomin_output)
mean(CUFA_nomin_utility_probs[,2], na.rm=TRUE)
mean(E_output[,2])
mean(CUFA_nomin_utility_probs[,2], na.rm=TRUE) - mean(E_output[,2])
#### Prepare data for GIS mapping ####
CUFA_150_prob_diffs <- data.frame(cell = E_output[ ,1], mean_prob_diff = CUFA_150_utility_probs[ ,2] - E_output[ ,2])
hist(CUFA_150_prob_diffs[ ,2])
CUFA_nomin_prob_diffs <- data.frame(cell = E_output[ ,1], mean_prob_diff = CUFA_nomin_utility_probs[ ,2] - E_output[ ,2])
hist(CUFA_nomin_prob_diffs[ ,2])
# Read x,y coordinate for the site
site5_xy <- read.table("data/coordinates/site5_coords_1000_to_4000.txt", header=TRUE)
# Trim x,y coordinate data so only cells with model outputs are included
site5_xy_trim <- site5_xy[site5_xy[ ,3] %in% E_output[ ,1],]
# Combine the coordinate and results data
CUFA_150_gis <- data.frame(site5_xy_trim,
existing_prob = E_output[, 2],
CUFA_150_prob = CUFA_150_utility_probs[, 2],
prob_diff = CUFA_150_prob_diffs[ ,2],
e_events = E_lengths[ ,2],
CUFA_150_events = CUFA_150_lengths[ ,2]
)
head(CUFA_150_gis)
write.table(CUFA_150_gis, file = "output/site5/CUFA_150_gis_site5.txt", sep = "\t", row.names = FALSE)
CUFA_nomin_gis <- data.frame(site5_xy_trim,
existing_prob = E_output[, 2],
CUFA_nomin_prob = CUFA_nomin_utility_probs[, 2],
prob_diff = CUFA_nomin_prob_diffs[ ,2],
e_events = E_lengths[ ,2],
CUFA_nomin_events = CUFA_nomin_lengths[ ,2]
)
head(CUFA_nomin_gis)
write.table(CUFA_nomin_gis, file = "output/site5/CUFA_nomin_gis_site5.txt", sep = "\t", row.names = FALSE)
# Create a spatial data frame by assigning spatial coordinates to the x,y columns
coordinates(CUFA_150_gis) <- c("x", "y")
coordinates(CUFA_nomin_gis) <- c("x", "y")
# X,Y coordinates are in New Mexico West state plane NAD83 (EPSG code 2904). This needs to be defined in the new object.
CUFA_150_gis@proj4string # Verify that no projection is currently specified
NMWSP <- CRS("+init=epsg:2904") # NMWSP stands for New Mexico West state plane
proj4string(CUFA_150_gis) <- NMWSP # Assign the proj4string slot the correct projection
CUFA_150_gis@proj4string # Verify that projection has been specified
proj4string(CUFA_nomin_gis) <- NMWSP # Assign the proj4string slot the correct projection
CUFA_nomin_gis@proj4string # Verify that projection has been specified
# Export data as shapefile
writeOGR(CUFA_150_gis, "/Users/Morrison/Documents/GIS/Projects/gila bn/shapefiles/site5", "CUFA_150_site5", driver="ESRI Shapefile", overwrite_layer=TRUE)
writeOGR(CUFA_nomin_gis, "/Users/Morrison/Documents/GIS/Projects/gila bn/shapefiles/site5", "CUFA_nomin_site5", driver="ESRI Shapefile", overwrite_layer=TRUE)
#### Combine all scenarios into single data frame for analysis ####
# Combine model ouput
E_site5 <- data.frame(combine.probs.E(), scenario = "existing")
CUFA_150_site5 <- data.frame(combine.probs.CUFA.150(), scenario = "CUFA_150")
CUFA_nomin_site5 <- data.frame(combine.probs.CUFA.nomin(), scenario = "CUFA_nomin")
combined_site5 <- data.frame(rbind(E_site5, CUFA_150_site5, CUFA_nomin_site5), site = "site5")
# Calculate and combine evidence in each cell
E_lengths_site5 <- combine.lengths.E()
CUFA_150_lengths_site5 <- combine.lengths.CUFA.150()
CUFA_nomin_lengths_site5 <- combine.lengths.CUFA.nomin()
# Calculate difference in events
E_lengthsdiff_site5 <- (E_lengths_site5[ ,1] - E_lengths_site5[ ,1])/E_lengths_site5[ ,1] * 100
CUFA_150_lengthsdiff_site5 <- (E_lengths_site5[ ,1] - CUFA_150_lengths_site5[ ,1])/E_lengths_site5[ ,1] * 100
CUFA_nomin_lengthsdiff_site5 <- (E_lengths_site5[ ,1] - CUFA_nomin_lengths_site5[ ,1])/E_lengths_site5[ ,1] * 100
# Apply utility function to calculate percent decrease in probabilities
E_utility_site5 <- utility_ftn(E_lengthsdiff_site5)
CUFA_150_utility_site5 <- utility_ftn(CUFA_150_lengthsdiff_site5)
CUFA_nomin_utility_site5 <- utility_ftn(CUFA_nomin_lengthsdiff_site5)
# Apply utility function decreases to probabilities
E_util_site5 <- E_site5[ ,1] * E_utility_site5
CUFA_150_util_site5 <- CUFA_150_site5[ ,1] * CUFA_150_utility_site5
CUFA_nomin_util_site5 <- CUFA_nomin_site5[ ,1] * CUFA_nomin_utility_site5
# Count the number of events in each month
E_all_events_site5 <- c(E_evidence1, E_evidence2, E_evidence3, E_evidence4, E_evidence5, E_evidence6, E_evidence7)
CUFA_150_all_events_site5 <- c(CUFA_150_evidence1, CUFA_150_evidence2, CUFA_150_evidence3, CUFA_150_evidence4, CUFA_150_evidence5, CUFA_150_evidence6, CUFA_150_evidence7)
CUFA_nomin_all_events_site5 <- c(CUFA_nomin_evidence1, CUFA_nomin_evidence2, CUFA_nomin_evidence3, CUFA_nomin_evidence4, CUFA_nomin_evidence5, CUFA_nomin_evidence6, CUFA_nomin_evidence7)
E_count_site5 <- countevents(E_all_events_site5)
CUFA_150_count_site5 <- countevents(CUFA_150_all_events_site5)
CUFA_nomin_count_site5 <- countevents(CUFA_nomin_all_events_site5)
# Make one large data frame with all the data
E_site5 <- data.frame(prob = E_site5$prob,
util_prob = E_util_site5,
prob_diff = E_util_site5 - E_site5$prob,
evidence = E_lengths_site5$evidence,
E_count_site5,
Q_bin = E_site5$Q_bin,
scenario = E_site5$scenario)
CUFA_150_site5 <- data.frame(prob = CUFA_150_site5$prob,
util_prob = CUFA_150_util_site5,
prob_diff = CUFA_150_util_site5 - E_site5$prob,
evidence = CUFA_150_lengths_site5$evidence,
CUFA_150_count_site5,
Q_bin = CUFA_150_site5$Q_bin,
scenario = CUFA_150_site5$scenario)
CUFA_nomin_site5 <- data.frame(prob = CUFA_nomin_site5$prob,
util_prob = CUFA_nomin_util_site5,
prob_diff = CUFA_nomin_util_site5 - E_site5$prob,
evidence = CUFA_nomin_lengths_site5$evidence,
CUFA_nomin_count_site5,
Q_bin = CUFA_nomin_site5$Q_bin,
scenario = CUFA_nomin_site5$scenario)
combined_site5 <- data.frame(rbind(E_site5, CUFA_150_site5, CUFA_nomin_site5), site = "site5")
# Export as .Rdata
save(combined_site5, file="output/combined_site5.Rdata")
# #### Approach #2 ####
#
# #### Account for cells that are not inundated and have zero probability ####
# CUFA_150_diffcells <- setdiff(E_output[,1], CUFA_150_output[,1]) # Finds cell numbers in existing that aren't in CUFA 150
# CUFA_nomin_diffcells <- setdiff(E_output[,1], CUFA_nomin_output[,1]) # Finds cell numbers in existing that aren't in CUFA no minimum
#
# #### Assign zero probabilities to cell numbers ####
# CUFA_150_zeroprobs <- data.frame(CUFA_150_diffcells, rep(0, times=length(CUFA_150_diffcells)))
# CUFA_nomin_zeroprobs <- data.frame(CUFA_nomin_diffcells, rep(0, times=length(CUFA_nomin_diffcells)))
# colnames(CUFA_150_zeroprobs) <- c("cell", "mean_prob")
# colnames(CUFA_nomin_zeroprobs) <- c("cell", "mean_prob")
#
# #### Combine model output to cells with zero probability ####
# CUFA_150_output_supp <- rbind(CUFA_150_output, CUFA_150_zeroprobs)
# CUFA_nomin_output_supp <- rbind(CUFA_nomin_output, CUFA_nomin_zeroprobs)
#
# #### Sort by cell number ####
# CUFA_150_output_supp <- CUFA_150_output_supp[order(CUFA_150_output_supp$cell),]
# CUFA_nomin_output_supp <- CUFA_nomin_output_supp[order(CUFA_nomin_output_supp$cell),]
#
# #### Assign zero lengths to cells ####
# CUFA_150_zerolengths <- data.frame(CUFA_150_diffcells, rep(0, times=length(CUFA_150_diffcells)))
# CUFA_nomin_zerolengths <- data.frame(CUFA_nomin_diffcells, rep(0, times=length(CUFA_nomin_diffcells)))
#
# colnames(CUFA_150_zerolengths) <- c("cell", "length")
# colnames(CUFA_nomin_zerolengths) <- c("cell", "length")
#
# #### Calculate the mean probability difference between existing conditons and each scenario ####
# CUFA_150_diff <- (CUFA_150_output_supp - E_output)*100
# mean(CUFA_150_diff[,2], na.rm=TRUE)
# CUFA_nomin_diff <- (CUFA_nomin_output_supp - E_output)*100
# mean(CUFA_nomin_diff[,2], na.rm=TRUE)
#
# #### Quick histogram plots of the probability differences ####
# hist(CUFA_150_diff[,2])
# hist(CUFA_nomin_diff[,2])
#
# #### Calculate the number of events (length) of each cell ####
# # CUFA 150
# CUFA_150_all_probs <- c(CUFA_150_q1_cell_probs, CUFA_150_q2_cell_probs, CUFA_150_q3_cell_probs, CUFA_150_q4_cell_probs, CUFA_150_q5_cell_probs, CUFA_150_q6_cell_probs) # Combine probabilities for all cells
# head(CUFA_150_all_probs) # Examine cell probs
# CUFA_150_lengths <- as.data.frame(sapply(CUFA_150_all_probs, length)) # Calculate the number of events in each cell and convert to data frame
# CUFA_150_lengths <- data.frame(c(as.numeric(row.names(CUFA_150_lengths))), CUFA_150_lengths[,1]) # Create new data frame with cell number in first column and number of events in second colume
# colnames(CUFA_150_lengths) <- c("cell", "length") # Name columns
# CUFA_150_lengths <- rbind(CUFA_150_lengths, CUFA_150_zerolengths) # Add cells to data frame that have zero length (no events)
# head(CUFA_150_lengths) # Preview the data table
# nrow(CUFA_150_lengths) # Count number of cells (rows)
# CUFA_150_mean <- mean(CUFA_150_lengths[,2]) # Calculate the mean number of events for all cells
# CUFA_150_mean # View mean
#
# # CUFA no minimum
# CUFA_nomin_all_probs <- c(CUFA_nomin_q1_cell_probs, CUFA_nomin_q2_cell_probs, CUFA_nomin_q3_cell_probs, CUFA_nomin_q4_cell_probs, CUFA_nomin_q5_cell_probs, CUFA_nomin_q6_cell_probs) # Combine probabilities for all cells
# head(CUFA_nomin_all_probs) # Examine cell probs
# CUFA_nomin_lengths <- as.data.frame(sapply(CUFA_nomin_all_probs, length)) # Calculate the number of events in each cell and convert to data frame
# CUFA_nomin_lengths <- data.frame(c(as.numeric(row.names(CUFA_nomin_lengths))), CUFA_nomin_lengths[,1]) # Create new data frame with cell number in first column and number of events in second colume
# colnames(CUFA_nomin_lengths) <- c("cell", "length") # Name columns
# CUFA_nomin_lengths <- rbind(CUFA_nomin_lengths, CUFA_nomin_zerolengths) # Add cells to data frame that have zero length (no events)
# head(CUFA_nomin_lengths) # Preview the data table
# nrow(CUFA_nomin_lengths) # Count number of cells (rows)
# CUFA_nomin_mean <- mean(CUFA_nomin_lengths[,2]) # Calculate the mean number of events for all cells
# CUFA_nomin_mean # View mean
#
#
# #### Calculate the difference in events between existing conditions and the scenario for each cell ####
# # Sort data frames according to cell number
# E_lengths <- E_lengths[order(E_lengths$cell),]
# CUFA_150_lengths <- CUFA_150_lengths[order(CUFA_150_lengths$cell),]
# CUFA_nomin_lengths <- CUFA_nomin_lengths[order(CUFA_nomin_lengths$cell),]
#
# # CUFA 150
# CUFA_150_pct_diff <- as.data.frame((E_lengths[,2]-CUFA_150_lengths[,2])/E_lengths[,2]*100)
# CUFA_150_pct_diff <- data.frame((CUFA_150_lengths[,1]), CUFA_150_pct_diff[,1])
# colnames(CUFA_150_pct_diff) <- c("cell","pct_decrease")
# head(CUFA_150_pct_diff)
# mean(CUFA_150_pct_diff[,2])
#
# # CUFA no minimum
# CUFA_nomin_pct_diff <- as.data.frame((E_lengths[,2]-CUFA_nomin_lengths[,2])/E_lengths[,2]*100)
# CUFA_nomin_pct_diff <- data.frame((CUFA_nomin_lengths[,1]), CUFA_nomin_pct_diff[,1])
# colnames(CUFA_nomin_pct_diff) <- c("cell","pct_decrease")
# head(CUFA_nomin_pct_diff)
# mean(CUFA_nomin_pct_diff[,2])
#
# #### Apply the utility function to each scenario ####
# # CUFA 150
# CUFA_150_utility <- utility_ftn(CUFA_150_pct_diff[,2])
# CUFA_150_utility <- cbind(CUFA_150_pct_diff[,1], CUFA_150_utility)
# colnames(CUFA_150_utility) <- c("cell","pct_of_network_output")
# head(CUFA_150_utility)
#
# CUFA_nomin_utility <- utility_ftn(CUFA_nomin_pct_diff[,2])
# CUFA_nomin_utility <- cbind(CUFA_nomin_pct_diff[,1], CUFA_nomin_utility)
# colnames(CUFA_nomin_utility) <- c("cell","pct_of_network_output")
# head(CUFA_nomin_utility)
#
# CUFA_150_utility_probs_supp <- CUFA_150_output_supp[,2] * CUFA_150_utility[,2]
# CUFA_150_utility_probs_supp <- cbind(CUFA_150_pct_diff[,1], CUFA_150_utility_probs_supp)
# colnames(CUFA_150_utility_probs_supp) <- c("cell","mean_prob")
# head(CUFA_150_utility_probs_supp)
# head(CUFA_150_output_supp)
# mean(CUFA_150_utility_probs_supp[,2], na.rm=TRUE)
# mean(E_output[,2])
#
# CUFA_nomin_utility_probs_supp <- CUFA_nomin_output_supp[,2] * CUFA_nomin_utility[,2]
# CUFA_nomin_utility_probs_supp <- cbind(CUFA_nomin_pct_diff[,1], CUFA_nomin_utility_probs_supp)
# colnames(CUFA_nomin_utility_probs_supp) <- c("cell","mean_prob")
# head(CUFA_nomin_utility_probs_supp)
# head(CUFA_nomin_output_supp)
# mean(CUFA_nomin_utility_probs_supp[,2], na.rm=TRUE)
# mean(E_output[,2])
#
# #### Calculate the relative difference between existing and scenarios on a cell-by-cell basis####
# CUFA_150_cell_diff <- (CUFA_150_utility_probs_supp[,2] - E_output[,2])/E_output[,2]*100
# CUFA_150_cell_diff <- cbind(CUFA_150_utility_probs_supp[,1], CUFA_150_cell_diff)
# colnames(CUFA_150_cell_diff) <- c("cell","relative_diff")
# head(CUFA_150_cell_diff)
#
# CUFA_nomin_cell_diff <- (CUFA_nomin_utility_probs_supp[,2] - E_output[,2])/E_output[,2]*100
# CUFA_nomin_cell_diff <- cbind(CUFA_nomin_utility_probs_supp[,1], CUFA_nomin_cell_diff)
# colnames(CUFA_nomin_cell_diff) <- c("cell","relative_diff")
# head(CUFA_nomin_cell_diff)
#
# E_evidence2[["2"]]
# CUFA_150_evidence2[["2"]]
# CUFA_nomin_evidence2[["2"]]
#
# E_evidence4[["15"]]
# CUFA_150_evidence4[["15"]]
# CUFA_nomin_evidence4[["15"]]