]> code.communitydata.science - ml_measurement_error_public.git/commitdiff
update simulations code
authorNathan TeBlunthuis <nathante@uw.edu>
Wed, 1 Mar 2023 00:13:36 +0000 (16:13 -0800)
committerNathan TeBlunthuis <nathante@uw.edu>
Wed, 1 Mar 2023 00:13:36 +0000 (16:13 -0800)
simulations/01_two_covariates.R
simulations/02_indep_differential.R
simulations/03_depvar.R
simulations/04_depvar_differential.R
simulations/Makefile
simulations/run_job.sbatch

index cd688c7d4b34d2456302299bc284cfedceb2c3f3..1f317be96eaa5ff9cd2803d06afde54e8d6f9f17 100644 (file)
@@ -79,6 +79,7 @@ parser <- add_argument(parser, "--Bzx", help='Effect of z on x', default=0.3)
 parser <- add_argument(parser, "--Bzy", help='Effect of z on y', default=-0.3)
 parser <- add_argument(parser, "--Bxy", help='Effect of x on y', default=0.3)
 parser <- add_argument(parser, "--Px", help='Base rate of x', default=0.5)
+parser <- add_argument(parser, "--confint_method", help='method for approximating confidence intervals', default='quad')
 
 args <- parse_args(parser)
 B0 <- 0
@@ -89,9 +90,9 @@ Bzx <- args$Bzx
 
 df <- simulate_data(args$N, args$m, B0, Bxy, Bzy, Bzx, Px, seed=args$seed + 500, y_explained_variance = args$y_explained_variance,  prediction_accuracy=args$prediction_accuracy)
 
-result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, Bzx=Bzx, 'Bzy'=Bzy, 'Px'=Px, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'accuracy_imbalance_difference'=args$accuracy_imbalance_difference, 'outcome_formula'=args$outcome_formula, 'proxy_formula'=args$proxy_formula,truth_formula=args$truth_formula, error='')
+result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, 'Bzx'=Bzx, 'Bzy'=Bzy, 'Px'=Px, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'accuracy_imbalance_difference'=args$accuracy_imbalance_difference, 'outcome_formula'=args$outcome_formula, 'proxy_formula'=args$proxy_formula,truth_formula=args$truth_formula, confint_method=args$confint_method,error='')
 
-outline <- run_simulation(df, result, outcome_formula=as.formula(args$outcome_formula), proxy_formula=as.formula(args$proxy_formula), truth_formula=as.formula(args$truth_formula))
+outline <- run_simulation(df, result, outcome_formula=as.formula(args$outcome_formula), proxy_formula=as.formula(args$proxy_formula), truth_formula=as.formula(args$truth_formula),confint_method=args$confint_method)
     
 outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE)
 if(file.exists(args$outfile)){
index 4e3a1324339856d0edf41d1a96c1a4ecb62fa3cb..9c33be717f0a2169c9eb2bd19d35204ea0a3fa53 100644 (file)
@@ -141,7 +141,7 @@ parser <- add_argument(parser, "--y_bias", help='coefficient of y on the probabi
 parser <- add_argument(parser, "--z_bias", help='coefficient of z on the probability a classification is correct', default=0)
 parser <- add_argument(parser, "--truth_formula", help='formula for the true variable', default="x~z")
 parser <- add_argument(parser, "--Px", help='base rate of x', default=0.5)
-
+parser <- add_argument(parser, "--confint_method", help='method for approximating confidence intervals', default='quad')
 args <- parse_args(parser)
 
 B0 <- 0
@@ -159,9 +159,9 @@ if(args$m < args$N){
     ## pc.df <- pc(suffStat=list(C=cor(df.pc),n=nrow(df.pc)),indepTest=gaussCItest,labels=names(df.pc),alpha=0.05)
     ## plot(pc.df)
 
-    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, 'Bzx'=args$Bzx, 'Bzy'=Bzy, 'Px'=Px, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'accuracy_imbalance_difference'=args$accuracy_imbalance_difference, 'y_bias'=args$y_bias,'outcome_formula'=args$outcome_formula, 'proxy_formula'=args$proxy_formula,truth_formula=args$truth_formula, error='')
+    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, 'Bzx'=args$Bzx, 'Bzy'=Bzy, 'Px'=Px, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'accuracy_imbalance_difference'=args$accuracy_imbalance_difference, 'y_bias'=args$y_bias,'outcome_formula'=args$outcome_formula, 'proxy_formula'=args$proxy_formula,truth_formula=args$truth_formula, confint_method=args$confint_method, error='')
 
-    outline <- run_simulation(df, result, outcome_formula=as.formula(args$outcome_formula), proxy_formula=as.formula(args$proxy_formula), truth_formula=as.formula(args$truth_formula))
+    outline <- run_simulation(df, result, outcome_formula=as.formula(args$outcome_formula), proxy_formula=as.formula(args$proxy_formula), truth_formula=as.formula(args$truth_formula),confint_method=args$confint_method)
     
    
  outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE)
index f0064f2940d93257b6ecfa481db7fa2b4f3c47ff..461c01a26f9251d446e3b41a8f06b6a60624e023 100644 (file)
@@ -79,6 +79,7 @@ parser <- add_argument(parser, "--Bzx", help='coeffficient of z on x', default=-
 parser <- add_argument(parser, "--B0", help='Base rate of y', default=0.5)
 parser <- add_argument(parser, "--outcome_formula", help='formula for the outcome variable', default="y~x+z")
 parser <- add_argument(parser, "--proxy_formula", help='formula for the proxy variable', default="w_pred~y")
+parser <- add_argument(parser, "--confint_method", help='method for getting confidence intervals', default="quad")
 
 args <- parse_args(parser)
 
@@ -91,9 +92,9 @@ if(args$m < args$N){
     df <- simulate_data(args$N, args$m, B0, Bxy, Bzy, Bzx, args$seed, args$prediction_accuracy)
 
 #    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'x_bias_y0'=args$x_bias_y0,'x_bias_y1'=args$x_bias_y1,'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula)
-    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy, 'Bzx'=Bzx,'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula)
+    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy, 'Bzx'=Bzx,'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula, 'confint_method'=args$confint_method)
 
-    outline <- run_simulation_depvar(df, result, outcome_formula = as.formula(args$outcome_formula), proxy_formula = as.formula(args$proxy_formula))
+    outline <- run_simulation_depvar(df, result, outcome_formula = as.formula(args$outcome_formula), proxy_formula = as.formula(args$proxy_formula), confint_method=args$confint_method)
 
     outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE)
 
index b3670807549249256da52cfc1fd0cb6ea8662b90..2c43f595bb5524b9a1d6f33bcaac8261f9f7d780 100644 (file)
@@ -31,12 +31,12 @@ source("simulation_base.R")
 
 ## one way to do it is by adding correlation to x.obs and y that isn't in w.
 ## in other words, the model is missing an important feature of x.obs that's related to y.
-simulate_data <- function(N, m, B0, Bxy, Bzy, seed, prediction_accuracy=0.73, z_bias=-0.75){
+simulate_data <- function(N, m, B0, Bxy, Bzy, Bxz=0, seed=0, prediction_accuracy=0.73, z_bias=-0.75){
     set.seed(seed)
 
     # make w and y dependent
     z <- rnorm(N,sd=0.5)
-    x <- rbinom(N,1,0.5)
+    x <- rbinom(N,1,plogis(Bxz*z))
 
     ystar <- Bzy * z + Bxy * x + B0
     y <- rbinom(N,1,plogis(ystar))
@@ -70,30 +70,32 @@ parser <- add_argument(parser, "--N", default=1000, help="number of observations
 parser <- add_argument(parser, "--m", default=500, help="m the number of ground truth observations")
 parser <- add_argument(parser, "--seed", default=17, help='seed for the rng')
 parser <- add_argument(parser, "--outfile", help='output file', default='example_4.feather')
-parser <- add_argument(parser, "--prediction_accuracy", help='how accurate is the predictive model?', default=0.79)
+parser <- add_argument(parser, "--prediction_accuracy", help='how accurate is the predictive model?', default=0.75)
 ## parser <- add_argument(parser, "--z_bias_y1", help='how is the classifier biased when y = 1?', default=-0.75)
 ## parser <- add_argument(parser, "--z_bias_y0", help='how is the classifier biased when y = 0 ?', default=0.75)
-parser <- add_argument(parser, "--z_bias", help='how is the classifier biased?', default=1.5)
-parser <- add_argument(parser, "--Bxy", help='coefficient of x on y', default=0.1)
-parser <- add_argument(parser, "--Bzy", help='coeffficient of z on y', default=-0.1)
-parser <- add_argument(parser, "--B0", help='coeffficient of z on y', default=-0.1)
+parser <- add_argument(parser, "--z_bias", help='how is the classifier biased?', default=-0.5)
+parser <- add_argument(parser, "--Bxy", help='coefficient of x on y', default=0.7)
+parser <- add_argument(parser, "--Bzy", help='coeffficient of z on y', default=-0.7)
+parser <- add_argument(parser, "--Bzx", help='coeffficient of z on y', default=1)
+parser <- add_argument(parser, "--B0", help='coeffficient of z on y', default=0)
 parser <- add_argument(parser, "--outcome_formula", help='formula for the outcome variable', default="y~x+z")
 parser <- add_argument(parser, "--proxy_formula", help='formula for the proxy variable', default="w_pred~y+z")
-
+parser <- add_argument(parser, "--confint_method", help='method for approximating confidence intervals', default='quad')
 args <- parse_args(parser)
 
 B0 <- args$B0
 Bxy <- args$Bxy
 Bzy <- args$Bzy
-
+Bzx <- args$Bzx
 
 if(args$m < args$N){
-    df <- simulate_data(args$N, args$m, B0, Bxy, Bzy, args$seed, args$prediction_accuracy, args$z_bias)
+    df <- simulate_data(args$N, args$m, B0, Bxy, Bzx, Bzy, args$seed, args$prediction_accuracy, args$z_bias)
 
-#    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'z_bias_y0'=args$z_bias_y0,'z_bias_y1'=args$z_bias_y1,'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula)
-    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'z_bias'=args$z_bias,'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula)
+#    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy,'Bzx'=Bzx, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'z_bias_y0'=args$z_bias_y0,'z_bias_y1'=args$z_bias_y1,'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula)
+    result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bzy'=Bzy,'Bzx'=Bzx, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'z_bias'=args$z_bias,'outcome_formula' = args$outcome_formula, 'proxy_formula' = args$proxy_formula, confint_method=args$confint_method)
 
-    outline <- run_simulation_depvar(df, result, outcome_formula = as.formula(args$outcome_formula), proxy_formula = as.formula(args$proxy_formula))
+    outline <- run_simulation_depvar(df, result, outcome_formula = as.formula(args$outcome_formula), proxy_formula = as.formula(args$proxy_formula),confint_method=args$confint_method)
+    print(outline$error.cor.z)
 
     outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE)
 
index 3e8fdd5c24a391171bf1f9e47de85140b9c96515..821280b52ddccd919787bc38b2ca385dc5a48917 100644 (file)
@@ -8,7 +8,7 @@ explained_variances=[0.1]
 
 all:main supplement
 main:remembr.RDS 
-supplement:robustness_1.RDS robustness_1_dv.RDS robustness_2.RDS robustness_2_dv.RDS robustness_3.RDS robustness_3_dv.RDS robustness_4.RDS robustness_4_dv.RDS 
+supplement:robustness_1.RDS robustness_1_dv.RDS robustness_2.RDS robustness_2_dv.RDS robustness_3.RDS robustness_3_dv.RDS robustness_3_proflik.RDS robustness_3_dv_proflik.RDS robustness_4.RDS robustness_4_dv.RDS 
 
 srun=sbatch --wait --verbose run_job.sbatch
 
@@ -30,7 +30,7 @@ example_1_jobs: 01_two_covariates.R simulation_base.R grid_sweep.py pl_methods.R
 example_1.feather: example_1_jobs 
        rm -f example_1.feather
        sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_1_jobs
-       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_1_jobs
+       sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_1_jobs
        sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_1_jobs
        sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_1_jobs
        sbatch --wait --verbose --array=4001-$(shell cat example_1_jobs | wc -l)  run_simulation.sbatch 0 example_1_jobs
@@ -41,10 +41,10 @@ example_2_jobs: 02_indep_differential.R simulation_base.R grid_sweep.py pl_metho
 example_2.feather: example_2_jobs 
        rm -f example_2.feather
        sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_2_jobs
-       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_2_jobs
        sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_2_jobs
        sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_2_jobs
-       sbatch --wait --verbose --array=4001-$(shell cat example_2_jobs | wc -l)  run_simulation.sbatch 0 example_2_jobs
+       sbatch --wait --verbose --array=4001-$(shell cat example_2_jobs | wc -l)
 
 
 # example_2_B_jobs: example_2_B.R
@@ -55,23 +55,24 @@ example_2.feather: example_2_jobs
 #      sbatch --wait --verbose --array=1-3000 run_simulation.sbatch 0 example_2_B_jobs
 
 example_3_jobs: 03_depvar.R simulation_base.R grid_sweep.py pl_methods.R
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 03_depvar.R" --arg_dict '{"N":${Ns},"m":${ms}, "Bxy":[0.7],"Bzy":[-0.7],"seed":${seeds}, "outfile":["example_3.feather"], "y_explained_variance":${explained_variances}}' --outfile example_3_jobs
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 03_depvar.R" --arg_dict '{"N":${Ns},"m":${ms}, "Bxy":[0.7],"Bzy":[-0.7],"Bzx":[1],"seed":${seeds}, "outfile":["example_3.feather"], "y_explained_variance":${explained_variances}}' --outfile example_3_jobs
 
 example_3.feather: example_3_jobs
        rm -f example_3.feather 
        sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_3_jobs
-       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_3_jobs
-       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_3_jobs
+       sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_3_jobs
+        sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_3_jobs
        sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_3_jobs
        sbatch --wait --verbose --array=4001-$(shell cat example_3_jobs | wc -l)  run_simulation.sbatch 0 example_3_jobs
 
 example_4_jobs: 04_depvar_differential.R simulation_base.R grid_sweep.py pl_methods.R
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 04_depvar_differential.R" --arg_dict '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[0.3], "prediction_accuracy":[0.73]}' --outfile example_4_jobs
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 04_depvar_differential.R" --arg_dict '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"Bzx":[1], "m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[-0.5], "prediction_accuracy":[0.73]}' --outfile example_4_jobs
 
 example_4.feather: example_4_jobs
        rm -f example_4.feather
        sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 example_4_jobs
        sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 example_4_jobs
+       sbatch --wait --verbose --array=2001-3001 run_simulation.sbatch 0 example_4_jobs
        sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 example_4_jobs
        sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 example_4_jobs
        sbatch --wait --verbose --array=4001-$(shell cat example_4_jobs | wc -l)  run_simulation.sbatch 0 example_4_jobs
@@ -86,63 +87,73 @@ remembr.RDS:example_1.feather example_2.feather example_3.feather example_4.feat
        ${srun} Rscript plot_dv_example.R --infile example_4.feather --name "plot.df.example.4"
 
 
-robustness_1_jobs: 02_indep_differential.R simulation_base.R grid_sweep.py
-       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_1.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~1"]}' --outfile robustness_1_jobs
-
-
-
 START=0
 STEP=1000
 ONE=1
 
-robustness_1.feather: robustness_1_jobs
-       $(eval END_1!=cat robustness_1_jobs | wc -l)
-       $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
-       rm -f robustness_1.feather
-       sbatch --wait --verbose --array=1-1000  run_simulation.sbatch 0 robustness_1_jobs
-       sbatch --wait --verbose --array=1001-2000  run_simulation.sbatch 0 robustness_1_jobs
-       sbatch --wait --verbose --array=2001-3000  run_simulation.sbatch 0 robustness_1_jobs
-       sbatch --wait --verbose --array=3001-4000  run_simulation.sbatch 0 robustness_1_jobs
-       sbatch --wait --verbose --array=4001-$(shell cat robustness_1_jobs | wc -l)  run_simulation.sbatch 0 robustness_1_jobs
+robustness_Ns=[1000,5000]
+robustness_robustness_ms=[100,200]
+
+#in robustness 1 / example 2 misclassification is correlated with Y.
+robustness_1_jobs_p1: 02_indep_differential.R simulation_base.R grid_sweep.py
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":[1000],"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_1.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3,0],"Bxy":[0.3],"Bzx":[1,0], "outcome_formula":["y~x+z"], "z_bias":[0, 0.5], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~1"]}' --outfile robustness_1_jobs_p1
 
-       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_jobs;)
+robustness_1_jobs_p2: 02_indep_differential.R simulation_base.R grid_sweep.py
+       sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":[5000],"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_1.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3,0],"Bxy":[0.3],"Bzx":[1,0], "outcome_formula":["y~x+z"], "z_bias":[0, 0.5], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~1"]}' --outfile robustness_1_jobs_p2
 
-robustness_1.RDS: robustness_1.feather
+robustness_1.feather: robustness_1_jobs_p1 robustness_1_jobs_p2
+       rm -f $@
+       $(eval END_1!=cat robustness_1_jobs_p1 | wc -l)
+       $(eval ITEROBUSTNESS_MS_1!=seq $(START) $(STEP) $(END_1))
+       $(eval END_2!=cat robustness_1_jobs_p2 | wc -l)
+       $(eval ITEROBUSTNESS_MS_2!=seq $(START) $(STEP) $(END_2))
+
+       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_jobs_p1;)
+       $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_jobs_p2;)
+
+robustness_1.RDS: robustness_1.feather summarize_estimator.R
        rm -f robustness_1.RDS
        ${srun} Rscript plot_example.R --infile $< --name "robustness_1" --remember-file $@
 
-robustness_1_dv_jobs: simulation_base.R 04_depvar_differential.R grid_sweep.py
-       ${srun} grid_sweep.py --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["robustness_1_dv.feather"], "proxy_formula":["w_pred~y"],"z_bias":[0.5]}' --outfile robustness_1_dv_jobs
+# when Bzy is 0 and zbias is not zero, we have the case where P(W|Y,X,Z) has an omitted variable that is conditionanlly independent from Y.  Note that X and Z are independent in this scenario.
+robustness_1_dv_jobs_p1: simulation_base.R 04_depvar_differential.R grid_sweep.py
+       ${srun} grid_sweep.py --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":[1000],"Bzx":[1], "Bxy":[0.7,0],"Bzy":[-0.7,0],"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_1_dv.feather"], "proxy_formula":["w_pred~y"],"z_bias":[-0.5]}' --outfile robustness_1_dv_jobs_p1
 
-robustness_1_dv.feather: robustness_1_dv_jobs
-       rm -f robustness_1_dv.feather
-       $(eval END_1!=cat robustness_1_dv_jobs | wc -l)
-       $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
-       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_dv_jobs;)
+robustness_1_dv_jobs_p2: simulation_base.R 04_depvar_differential.R grid_sweep.py
+       ${srun} grid_sweep.py --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":[5000],"Bzx":[1], "Bxy":[0.7,0],"Bzy":[-0.7,0],"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_1_dv.feather"], "proxy_formula":["w_pred~y"],"z_bias":[-0.5]}' --outfile robustness_1_dv_jobs_p2
 
+robustness_1_dv.feather: robustness_1_dv_jobs_p1 robustness_1_dv_jobs_p2
+       rm -f $@
+       $(eval END_1!=cat robustness_1_dv_jobs_p1 | wc -l)
+       $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
+       $(eval END_2!=cat robustness_1_dv_jobs_p2 | wc -l)
+       $(eval ITEMS_2!=seq $(START) $(STEP) $(END_1))
+       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_dv_jobs_p1;)
+       $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_1_dv_jobs_p2;)
 
-robustness_1_dv.RDS: robustness_1_dv.feather
+robustness_1_dv.RDS: robustness_1_dv.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript plot_dv_example.R --infile $< --name "robustness_1_dv" --remember-file $@
 
 
 robustness_2_jobs_p1: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.60,0.65]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.60,0.65]}' --outfile $@
 
 robustness_2_jobs_p2: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.70,0.75]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.70,0.75]}' --outfile $@
 
 robustness_2_jobs_p3: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.80,0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.80,0.85]}' --outfile $@
 
 robustness_2_jobs_p4: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.90,0.95]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.90,0.95]}' --outfile $@
 
 robustness_2.feather: robustness_2_jobs_p1 robustness_2_jobs_p2 robustness_2_jobs_p3 robustness_2_jobs_p4
+       rm $@
        $(eval END_1!=cat robustness_2_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_2_jobs_p2 | wc -l)
@@ -157,27 +168,28 @@ robustness_2.feather: robustness_2_jobs_p1 robustness_2_jobs_p2 robustness_2_job
        $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_2_jobs_p3;)
        $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_2_jobs_p4;)
 
-robustness_2.RDS: plot_example.R robustness_2.feather 
+robustness_2.RDS: plot_example.R robustness_2.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_2" --remember-file $@
 
 robustness_2_dv_jobs_p1: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.60,0.65]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.60,0.65]}' --outfile $@
 
 robustness_2_dv_jobs_p2: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.70,0.75]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.70,0.75]}' --outfile $@
 
 robustness_2_dv_jobs_p3: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.80,0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.80,0.85]}' --outfile $@
 
 robustness_2_dv_jobs_p4: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.90,0.95]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_2_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.90,0.95]}' --outfile $@
 
 robustness_2_dv.feather: robustness_2_dv_jobs_p1 robustness_2_dv_jobs_p2 robustness_2_dv_jobs_p3 robustness_2_dv_jobs_p4
+       rm -f $@
        $(eval END_1!=cat robustness_2_dv_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_2_dv_jobs_p2 | wc -l)
@@ -192,24 +204,40 @@ robustness_2_dv.feather: robustness_2_dv_jobs_p1 robustness_2_dv_jobs_p2 robustn
        $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_2_dv_jobs_p3;)
        $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_2_dv_jobs_p4;)
 
-robustness_2_dv.RDS: plot_example.R robustness_2_dv.feather 
+robustness_2_dv.RDS: plot_dv_example.R robustness_2_dv.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_2_dv" --remember-file $@
 
 
+robustness_3_proflik_jobs: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
+       rm -f $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":[1000],"m":[100], "seed":${seeds}, "outfile":["robustness_3_proflik.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"Px":[0.5,0.6,0.7,0.8,0.9,0.95], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85], "confint_method":['spline']}' --outfile $@
+
+robustness_3_proflik.feather: robustness_3_proflik_jobs
+       rm -f $@
+       $(eval END_1!=cat robustness_3_proflik_jobs | wc -l)
+       $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
+       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_3_proflik_jobs;)
+
+robustness_3_proflik.RDS: plot_example.R robustness_3_proflik.feather summarize_estimator.R
+       rm -f $@
+       ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_3_proflik" --remember-file $@
+
+
 robustness_3_jobs_p1: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3],"Px":[0.5,0.6], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"Px":[0.5,0.6], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 robustness_3_jobs_p2: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3],"Px":[0.7,0.8], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"Px":[0.7,0.8], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 robustness_3_jobs_p3: grid_sweep.py 01_two_covariates.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3],"Px":[0.9,0.95], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 01_two_covariates.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"Px":[0.9,0.95], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 robustness_3.feather: robustness_3_jobs_p1 robustness_3_jobs_p2 robustness_3_jobs_p3
+       rm -f $@
        $(eval END_1!=cat robustness_3_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_3_jobs_p2 | wc -l)
@@ -221,26 +249,42 @@ robustness_3.feather: robustness_3_jobs_p1 robustness_3_jobs_p2 robustness_3_job
        $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_3_jobs_p2;)
        $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_3_jobs_p3;)
 
-robustness_3.RDS: plot_example.R robustness_3.feather 
+robustness_3.RDS: plot_example.R robustness_3.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_3" --remember-file $@
 
-robustness_3_dv_jobs_p1: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
+robustness_3_dv_proflik_jobs: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
+       rm -f $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":[1000],"m":[100], "seed":${seeds}, "outfile":["robustness_3_dv_proflik.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"B0":[0,0.405,0.846,1.386,2.197,2.944], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"confint_method":['spline']}' --outfile $@
+
+robustness_3_dv_proflik.feather: robustness_3_dv_proflik_jobs
+       rm -f $@
+       $(eval END_1!=cat robustness_3_dv_proflik_jobs | wc -l)
+       $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
+       $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_3_dv_proflik_jobs;)
+
+robustness_3_dv_proflik.RDS: plot_dv_example.R robustness_3_dv_proflik.feather summarize_estimator.R
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3],"B0":[0.5,0.6], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_3_dv_proflik" --remember-file $@
+
+
+ robustness_3_dv_jobs_p1: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
+       rm -f $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"B0":[0,0.405], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 
 
 robustness_3_dv_jobs_p2: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3],"B0":[0.7,0.8], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1],"B0":[0.847,1.386], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 
 robustness_3_dv_jobs_p3: grid_sweep.py 03_depvar.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "B0":[0.9,0.95], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 03_depvar.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_3_dv.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "B0":[2.197,2.944], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85]}' --outfile $@
 
 robustness_3_dv.feather: robustness_3_dv_jobs_p1 robustness_3_dv_jobs_p2 robustness_3_dv_jobs_p3
+       rm -f $@
        $(eval END_1!=cat robustness_3_dv_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_3_dv_jobs_p2 | wc -l)
@@ -253,28 +297,26 @@ robustness_3_dv.feather: robustness_3_dv_jobs_p1 robustness_3_dv_jobs_p2 robustn
         $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_3_dv_jobs_p3;)
 
 
-robustness_3_dv.RDS: plot_dv_example.R robustness_3_dv.feather 
+robustness_3_dv.RDS: plot_dv_example.R robustness_3_dv.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_3_dv" --remember-file $@
 
 
+
 robustness_4_jobs_p1: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85],"y_bias":[-1,-0.85]}' --outfile $@
+       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85],"y_bias":[-2.944,-2.197]}' --outfile $@
 
 robustness_4_jobs_p2: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85], "y_bias":[-0.70,-0.55]}' --outfile $@
+       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85], "y_bias":[-1.386,-0.846]}' --outfile $@
 
 robustness_4_jobs_p3: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85],"y_bias":[-0.4,-0.25]}' --outfile $@
-
-robustness_4_jobs_p4: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py
-       rm -f $@
-       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85],"y_bias":[-0.1,0]}' --outfile $@
+       ${srun} $< --command 'Rscript 02_indep_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4.feather"],"y_explained_variance":${explained_variances},  "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~z"], "prediction_accuracy":[0.85],"y_bias":[-0.405,-0.25]}' --outfile $@
 
 robustness_4.feather: robustness_4_jobs_p1 robustness_4_jobs_p2 robustness_4_jobs_p3
+       rm -f $@
        $(eval END_1!=cat robustness_4_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_4_jobs_p2 | wc -l)
@@ -286,48 +328,52 @@ robustness_4.feather: robustness_4_jobs_p1 robustness_4_jobs_p2 robustness_4_job
        $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_jobs_p2;)
        $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_jobs_p3;)
 
-robustness_4.RDS: plot_example.R robustness_4.feather 
+robustness_4.RDS: plot_example.R robustness_4.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_4" --remember-file $@
 
 
-# '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[0.5]}' --outfile example_4_jobs
+# '{"N":${robustness_Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[0.5]}' --example_4_jobs
 
 robustness_4_dv_jobs_p1: grid_sweep.py 04_depvar_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0,0.1]}' --outfile $@
+       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0,0.1]}' --outfile $@
 
 robustness_4_dv_jobs_p2: grid_sweep.py 04_depvar_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0.25,0.4]}' --outfile $@
+       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0.25,0.405]}' --outfile $@
 
 robustness_4_dv_jobs_p3: grid_sweep.py 04_depvar_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0.55,0.7]}' --outfile $@
+       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[1],"outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0.846,1.386]}' --outfile $@
 
 robustness_4_dv_jobs_p4: grid_sweep.py 04_depvar_differential.R simulation_base.R grid_sweep.py
        rm -f $@
-       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"],"Bzy":[-0.7],"Bxy":[0.7], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[0.85,1]}' --outfile $@
+       ${srun} $< --command 'Rscript 04_depvar_differential.R'  --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_4_dv.feather"],"Bzy":[-0.7],"Bxy":[0.7],"Bzx":[1], "outcome_formula":["y~x+z"], "prediction_accuracy":[0.85],"z_bias":[2.197,2.944]}' --outfile $@
 
-robustness_4_dv.feather: robustness_4_dv_jobs_p1 robustness_4_dv_jobs_p2 robustness_4_dv_jobs_p3
+robustness_4_dv.feather: robustness_4_dv_jobs_p1 robustness_4_dv_jobs_p2 robustness_4_dv_jobs_p3 robustness_4_dv_jobs_p4
+       rm -f $@
        $(eval END_1!=cat robustness_4_dv_jobs_p1 | wc -l)
        $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1))
        $(eval END_2!=cat robustness_4_dv_p2 | wc -l)
        $(eval ITEMS_2!=seq $(START) $(STEP) $(END_2))
        $(eval END_3!=cat robustness_4_dv_p3 | wc -l)
        $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3))
+       $(eval END_3!=cat robustness_4_dv_p4 | wc -l)
+       $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3))
+
 
        $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_dv_jobs_p1;)
        $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_dv_jobs_p2;)
        $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_dv_jobs_p3;)
+       $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_dv_jobs_p4;)
 
 
-robustness_4_dv.RDS: plot_dv_example.R robustness_4_dv.feather 
+robustness_4_dv.RDS: plot_dv_example.R robustness_4_dv.feather summarize_estimator.R
        rm -f $@
        ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_4" --remember-file $@
 
 
-
 clean_main:
        rm -f remembr.RDS
        rm -f example_1_jobs
@@ -359,5 +405,4 @@ clean_all:
 #      sbatch --wait --verbose --array=3001-6001 run_simulation.sbatch 0 example_2_B_mecor_jobs
 
 
-
 .PHONY: supplement
index 856178879a626a972c2a318da3420b7d0b56d184..591b1787755ead24ad18837856116c51e04056f8 100644 (file)
@@ -1,8 +1,8 @@
 #!/bin/bash
 #SBATCH --job-name="simulate measurement error models"
 ## Allocation Definition
-#SBATCH --account=comdata
-#SBATCH --partition=compute-bigmem
+#SBATCH --account=comdata-ckpt
+#SBATCH --partition=ckpt
 ## Resources
 #SBATCH --nodes=1    
 ## Walltime (4 hours)
@@ -14,4 +14,5 @@
 #SBATCH --chdir /gscratch/comdata/users/nathante/ml_measurement_error_public/simulations
 #SBATCH --output=simulation_jobs/%A_%a.out
 #SBATCH --error=simulation_jobs/%A_%a.err
+echo "$@"
 "$@"

Community Data Science Collective || Want to submit a patch?