-
Notifications
You must be signed in to change notification settings - Fork 1
/
compute_bias_all_sims.R
545 lines (453 loc) · 45.4 KB
/
compute_bias_all_sims.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
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
source('./src/dependencies.R')
source('./src/estimation_functions.R')
#################################################################################################
# Set sample size for estimation sample used to compute the conditional bias
n <- 5000
reps <- 200
#################################################################################################
load("./params/simulation_parameters_continuous_2,8,32,64,128.RData")
with(simulation_parameters_continuous, {
prob_trt1_stacked_classifier <- prob_outcome1_trt1_stacked_classifier <- prob_outcome1_trt0_stacked_classifier <-
prob_trt1_boosted_tree <- prob_outcome1_trt1_boosted_tree <- prob_outcome1_trt0_boosted_tree <-
prob_trt1_random_forest <- prob_outcome1_trt1_random_forest <- prob_outcome1_trt0_random_forest <-
prob_trt1_knn <- prob_outcome1_trt1_knn <- prob_outcome1_trt0_knn <-
prob_trt1_lasso <- prob_outcome1_trt1_lasso <- prob_outcome1_trt0_lasso <-
prob_trt1_glm <- prob_outcome1_trt1_glm <- prob_outcome1_trt0_glm <-
p_X <- p_Y <- X <- Y <- vector()
for(rep in 1:reps) {
Z1 = runif(n, 0, 1); Z2 = runif(n, 0, 1); Z3 = runif(n, 0, 1); Z4 = runif(n, 0, 1); Z5 = runif(n, 0, 1)
Z6 = runif(n, 0, 1); Z7 = runif(n, 0, 1); Z8 = runif(n, 0, 1); Z9 = runif(n, 0, 1); Z10 = runif(n, 0, 1)
sim_data_est <- create_b_spline_basis(
data=data.frame(Z1),
continuous_vars=c('Z1'),
knots=true_knots,
boundary_knots=list(c(0,1)),
polynomial_degree=3,
degree_of_interactions=1
)
p_X[((n*(rep-1))+1):(n*(rep))] = sim_data_est %*% theta_X
p_Y[((n*(rep-1))+1):(n*(rep))] = sim_data_est %*% theta_Y
X[((n*(rep-1))+1):(n*(rep))] = rbinom(n, 1, p_X[((n*(rep-1))+1):(n*(rep))])
Y[((n*(rep-1))+1):(n*(rep))] = rbinom(n, 1, p_Y[((n*(rep-1))+1):(n*(rep))])
sim_data_est <- data.frame(Z1=Z1, Z2=Z2, Z3=Z3, Z4=Z4, Z5=Z5, Z6=Z6, Z7=Z7, Z8=Z8, Z9=Z9, Z10=Z10,
X=X[((n*(rep-1))+1):(n*(rep))],
Y=Y[((n*(rep-1))+1):(n*(rep))])
prob_trt1_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_trt,
params_random_forest = params_random_forest_trt,
params_knn = params_knn_trt,
params_lasso = params_lasso_trt,
params_glm = params_glm_trt,
meta_model = meta_model_trt$meta_model,
meta_model_formula = meta_model_trt$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_outcome,
params_random_forest = params_random_forest_outcome,
params_knn = params_knn_outcome,
params_lasso = params_lasso_outcome,
params_glm = params_glm_outcome,
meta_model = meta_model_outcome$meta_model,
meta_model_formula = meta_model_outcome$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_outcome,
params_random_forest = params_random_forest_outcome,
params_knn = params_knn_outcome,
params_lasso = params_lasso_outcome,
params_glm = params_glm_outcome,
meta_model = meta_model_outcome$meta_model,
meta_model_formula = meta_model_outcome$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_boosted_tree_trt,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_boosted_tree_outcome,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_boosted_tree_outcome,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_random_forest_trt,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_random_forest_outcome,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_random_forest_outcome,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_knn_trt,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_knn_outcome,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_knn_outcome,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_lasso_trt,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_lasso_outcome,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_lasso_outcome,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_glm_trt,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_glm_outcome,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_glm_outcome,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
}
bias_stacked <- compute_bias(prob_trt1_est = prob_trt1_stacked_classifier, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_stacked_classifier,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_stacked_classifier,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_boosted_tree <- compute_bias(prob_trt1_est = prob_trt1_boosted_tree, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_boosted_tree,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_boosted_tree,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_random_forest <- compute_bias(prob_trt1_est = prob_trt1_random_forest, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_random_forest,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_random_forest,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_knn <- compute_bias(prob_trt1_est = prob_trt1_knn, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_knn,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_knn,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_lasso <- compute_bias(prob_trt1_est = prob_trt1_lasso, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_lasso,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_lasso,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_glm <- compute_bias(prob_trt1_est = prob_trt1_glm, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_glm,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_glm,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_out <- list(bias_stacked = bias_stacked,
bias_boosted_tree = bias_boosted_tree,
bias_random_forest = bias_random_forest,
bias_knn = bias_knn,
bias_lasso = bias_lasso,
bias_glm = bias_glm)
cs_bias_stacked <- compute_CS_bias(prob_trt1_est = prob_trt1_stacked_classifier, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_stacked_classifier,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_stacked_classifier,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_boosted_tree <- compute_CS_bias(prob_trt1_est = prob_trt1_boosted_tree, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_boosted_tree,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_boosted_tree,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_random_forest <- compute_CS_bias(prob_trt1_est = prob_trt1_random_forest, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_random_forest,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_random_forest,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_knn <- compute_CS_bias(prob_trt1_est = prob_trt1_knn, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_knn,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_knn,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_lasso <- compute_CS_bias(prob_trt1_est = prob_trt1_lasso, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_lasso,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_lasso,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_glm <- compute_CS_bias(prob_trt1_est = prob_trt1_glm, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_glm,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_glm,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_out <- list(cs_bias_stacked = cs_bias_stacked,
cs_bias_boosted_tree = cs_bias_boosted_tree,
cs_bias_random_forest = cs_bias_random_forest,
cs_bias_knn = cs_bias_knn,
cs_bias_lasso = cs_bias_lasso,
cs_bias_glm = cs_bias_glm)
save(bias_out,
file=paste0("./output/bias_continuous.RData"))
save(cs_bias_out,
file=paste0("./output/cs_bias_continuous.RData"))
})
#################################################################################################
load("./params/simulation_parameters_binary_1,2,3,4,5,6,7,8.RData")
with(simulation_parameters_binary, {
prob_trt1_stacked_classifier <- prob_outcome1_trt1_stacked_classifier <- prob_outcome1_trt0_stacked_classifier <-
prob_trt1_boosted_tree <- prob_outcome1_trt1_boosted_tree <- prob_outcome1_trt0_boosted_tree <-
prob_trt1_random_forest <- prob_outcome1_trt1_random_forest <- prob_outcome1_trt0_random_forest <-
prob_trt1_knn <- prob_outcome1_trt1_knn <- prob_outcome1_trt0_knn <-
prob_trt1_lasso <- prob_outcome1_trt1_lasso <- prob_outcome1_trt0_lasso <-
prob_trt1_glm <- prob_outcome1_trt1_glm <- prob_outcome1_trt0_glm <-
p_X <- p_Y <- X <- Y <- vector()
for(rep in 1:reps) {
sim_data_est <- data.frame(
Z1 = as.numeric(rbernoulli(n,p=0.5)), Z2 = as.numeric(rbernoulli(n,p=0.5)),
Z3 = as.numeric(rbernoulli(n,p=0.5)), Z4 = as.numeric(rbernoulli(n,p=0.5)),
Z5 = as.numeric(rbernoulli(n,p=0.5)), Z6 = as.numeric(rbernoulli(n,p=0.5)),
Z7 = as.numeric(rbernoulli(n,p=0.5)), Z8 = as.numeric(rbernoulli(n,p=0.5)),
Z9 = as.numeric(rbernoulli(n,p=0.5)), Z10 = as.numeric(rbernoulli(n,p=0.5))
) %>%
merge(., key_trt) %>%
merge(., key_outcome) %>%
mutate(
X = rbinom(n(), 1, val_trt),
Y = rbinom(n(), 1, val_outcome)
)
p_X[((n*(rep-1))+1):(n*(rep))] <- sim_data_est$val_trt
p_Y[((n*(rep-1))+1):(n*(rep))] <- sim_data_est$val_outcome
X[((n*(rep-1))+1):(n*(rep))] <- sim_data_est$X
Y[((n*(rep-1))+1):(n*(rep))] <- sim_data_est$Y
sim_data_est <- sim_data_est %>% dplyr::select(-c(val_trt, val_outcome))
prob_trt1_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_trt,
params_random_forest = params_random_forest_trt,
params_knn = params_knn_trt,
params_lasso = params_lasso_trt,
params_glm = params_glm_trt,
meta_model = meta_model_trt$meta_model,
meta_model_formula = meta_model_trt$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_outcome,
params_random_forest = params_random_forest_outcome,
params_knn = params_knn_outcome,
params_lasso = params_lasso_outcome,
params_glm = params_glm_outcome,
meta_model = meta_model_outcome$meta_model,
meta_model_formula = meta_model_outcome$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_stacked_classifier[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_stacked_classifier(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params_boosted_tree = params_boosted_tree_outcome,
params_random_forest = params_random_forest_outcome,
params_knn = params_knn_outcome,
params_lasso = params_lasso_outcome,
params_glm = params_glm_outcome,
meta_model = meta_model_outcome$meta_model,
meta_model_formula = meta_model_outcome$meta_model_formula,
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_boosted_tree_trt,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_boosted_tree_outcome,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_boosted_tree[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_boosted_tree_outcome,
model = "boosted_tree",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_random_forest_trt,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_random_forest_outcome,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_random_forest[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_random_forest_outcome,
model = "random_forest",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_knn_trt,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_knn_outcome,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_knn[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_knn_outcome,
model = "knn",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_lasso_trt,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_lasso_outcome,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_lasso[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_lasso_outcome,
model = "lasso",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
prob_trt1_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y, X)),
label_vector = sim_data_tr %>% dplyr::select(X) %>% {.[[1]]},
params = params_glm_trt,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y, X)))
prob_outcome1_trt1_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_glm_outcome,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 1))
prob_outcome1_trt0_glm[((n*(rep-1))+1):(n*(rep))] <- estimate_prob_individual_model(covariates_df = sim_data_tr %>% dplyr::select(-c(Y)),
label_vector = sim_data_tr %>% dplyr::select(Y) %>% {.[[1]]},
params = params_glm_outcome,
model = "glm",
predict_data = sim_data_est %>% dplyr::select(-c(Y)) %>% mutate(X = 0))
}
bias_stacked <- compute_bias(prob_trt1_est = prob_trt1_stacked_classifier, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_stacked_classifier,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_stacked_classifier,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_boosted_tree <- compute_bias(prob_trt1_est = prob_trt1_boosted_tree, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_boosted_tree,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_boosted_tree,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_random_forest <- compute_bias(prob_trt1_est = prob_trt1_random_forest, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_random_forest,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_random_forest,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_knn <- compute_bias(prob_trt1_est = prob_trt1_knn, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_knn,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_knn,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_lasso <- compute_bias(prob_trt1_est = prob_trt1_lasso, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_lasso,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_lasso,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_glm <- compute_bias(prob_trt1_est = prob_trt1_glm, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_glm,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_glm,
prob_outcome1_trt0_true = p_Y,
trt = X)
bias_out <- list(bias_stacked = bias_stacked,
bias_boosted_tree = bias_boosted_tree,
bias_random_forest = bias_random_forest,
bias_knn = bias_knn,
bias_lasso = bias_lasso,
bias_glm = bias_glm)
cs_bias_stacked <- compute_CS_bias(prob_trt1_est = prob_trt1_stacked_classifier, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_stacked_classifier,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_stacked_classifier,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_boosted_tree <- compute_CS_bias(prob_trt1_est = prob_trt1_boosted_tree, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_boosted_tree,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_boosted_tree,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_random_forest <- compute_CS_bias(prob_trt1_est = prob_trt1_random_forest, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_random_forest,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_random_forest,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_knn <- compute_CS_bias(prob_trt1_est = prob_trt1_knn, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_knn,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_knn,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_lasso <- compute_CS_bias(prob_trt1_est = prob_trt1_lasso, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_lasso,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_lasso,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_glm <- compute_CS_bias(prob_trt1_est = prob_trt1_glm, prob_trt1_true = p_X,
prob_outcome1_trt1_est = prob_outcome1_trt1_glm,
prob_outcome1_trt1_true = p_Y,
prob_outcome1_trt0_est = prob_outcome1_trt0_glm,
prob_outcome1_trt0_true = p_Y,
trt = X)
cs_bias_out <- list(cs_bias_stacked = cs_bias_stacked,
cs_bias_boosted_tree = cs_bias_boosted_tree,
cs_bias_random_forest = cs_bias_random_forest,
cs_bias_knn = cs_bias_knn,
cs_bias_lasso = cs_bias_lasso,
cs_bias_glm = cs_bias_glm)
save(bias_out,
file=paste0("./output/bias_binary.RData"))
save(cs_bias_out,
file=paste0("./output/cs_bias_binary.RData"))
})