From: Nathan TeBlunthuis Date: Fri, 8 Sep 2023 16:01:31 +0000 (-0700) Subject: changes from klone X-Git-Url: https://code.communitydata.science/ml_measurement_error_public.git/commitdiff_plain/214551f74cc94ef1fa2f24aa317265c25bb03757?ds=inline changes from klone --- diff --git a/simulations/03_depvar.R b/simulations/03_depvar.R index 461c01a..ec1e231 100644 --- a/simulations/03_depvar.R +++ b/simulations/03_depvar.R @@ -73,8 +73,8 @@ parser <- add_argument(parser, "--y_explained_variance", help='what proportion o parser <- add_argument(parser, "--prediction_accuracy", help='how accurate is the predictive model?', default=0.72) ## parser <- add_argument(parser, "--x_bias_y1", help='how is the classifier biased when y = 1?', default=-0.75) ## parser <- add_argument(parser, "--x_bias_y0", help='how is the classifier biased when y = 0 ?', default=0.75) -parser <- add_argument(parser, "--Bxy", help='coefficient of x on y', default=0.01) -parser <- add_argument(parser, "--Bzy", help='coeffficient of z on y', default=-0.01) +parser <- add_argument(parser, "--Bxy", help='coefficient of x on y', default=0.3) +parser <- add_argument(parser, "--Bzy", help='coeffficient of z on y', default=-0.3) parser <- add_argument(parser, "--Bzx", help='coeffficient of z on x', default=-0.5) 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") diff --git a/simulations/03_indep_differential_nonnorm.R b/simulations/03_indep_differential_nonnorm.R new file mode 100644 index 0000000..de3346e --- /dev/null +++ b/simulations/03_indep_differential_nonnorm.R @@ -0,0 +1,185 @@ +### EXAMPLE 1: demonstrates how measurement error can lead to a type sign error in a covariate +### What kind of data invalidates fong + tyler? +### Even when you have a good predictor, if it's biased against a covariate you can get the wrong sign. +### Even when you include the proxy variable in the regression. +### But with some ground truth and multiple imputation, you can fix it. + +library(argparser) +library(mecor) +library(ggplot2) +library(data.table) +library(filelock) +library(arrow) +library(Amelia) +library(Zelig) +library(predictionError) +options(amelia.parallel="no", + amelia.ncpus=1) +setDTthreads(40) + +source("simulation_base.R") + +## SETUP: +### we want to estimate x -> y; x is MAR +### we have x -> k; k -> w; x -> w is used to predict x via the model w. +### A realistic scenario is that we have an NLP model predicting something like "racial harassment" in social media comments +### The labels x are binary, but the model provides a continuous predictor + +### simulation: +#### how much power do we get from the model in the first place? (sweeping N and m) +#### + +## 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, Bzx, Bzy, seed, y_explained_variance=0.025, prediction_accuracy=0.73, y_bias=-0.8,z_bias=0,Px=0.5,accuracy_imbalance_difference=0.3,sd_y_mixin=1){ + set.seed(seed) + # make w and y dependent + z <- rnorm(N,sd=0.5) + x <- rbinom(N, 1, plogis(Bzx * z + qlogis(Px))) + ## following Fong + Tyler: mix y with a Bernoulli(0.15) × |N (0, 20)| to make a skewed non-normal distribution + y.var.epsilon <- (var(Bzy * z) + var(Bxy *x) + 2*cov(Bzy*z,Bxy*x)) * ((1-y_explained_variance)/y_explained_variance) + y.epsilon <- rnorm(N, sd = sqrt(y.var.epsilon)) + y <- Bzy * z + Bxy * x + y.epsilon + rbinom(N,1,0.15) * rnorm(N,0,sd_y_mixin) + + df <- data.table(x=x,y=y,z=z) + + if(m < N){ + df <- df[sample(nrow(df), m), x.obs := x] + } else { + df <- df[, x.obs := x] + } + + ## probablity of an error is correlated with y + ## pz <- mean(z) + ## accuracy_imbalance_ratio <- (prediction_accuracy + accuracy_imbalance_difference/2) / (prediction_accuracy - accuracy_imbalance_difference/2) + + ## # this works because of conditional probability + ## accuracy_z0 <- prediction_accuracy / (pz*(accuracy_imbalance_ratio) + (1-pz)) + ## accuracy_z1 <- accuracy_imbalance_ratio * accuracy_z0 + + ## z0x0 <- df[(z==0) & (x==0)]$x + ## z0x1 <- df[(z==0) & (x==1)]$x + ## z1x0 <- df[(z==1) & (x==0)]$x + ## z1x1 <- df[(z==1) & (x==1)]$x + + ## yz0x0 <- df[(z==0) & (x==0)]$y + ## yz0x1 <- df[(z==0) & (x==1)]$y + ## yz1x0 <- df[(z==1) & (x==0)]$y + ## yz1x1 <- df[(z==1) & (x==1)]$y + + ## nz0x0 <- nrow(df[(z==0) & (x==0)]) + ## nz0x1 <- nrow(df[(z==0) & (x==1)]) + ## nz1x0 <- nrow(df[(z==1) & (x==0)]) + ## nz1x1 <- nrow(df[(z==1) & (x==1)]) + + ## yz1 <- df[z==1]$y + ## yz1 <- df[z==1]$y + + ## # tranform yz0.1 into a logistic distribution with mean accuracy_z0 + ## acc.z0x0 <- plogis(0.5*scale(yz0x0) + qlogis(accuracy_z0)) + ## acc.z0x1 <- plogis(0.5*scale(yz0x1) + qlogis(accuracy_z0)) + ## acc.z1x0 <- plogis(1.5*scale(yz1x0) + qlogis(accuracy_z1)) + ## acc.z1x1 <- plogis(1.5*scale(yz1x1) + qlogis(accuracy_z1)) + + ## w0z0x0 <- (1-z0x0)**2 + (-1)**(1-z0x0) * acc.z0x0 + ## w0z0x1 <- (1-z0x1)**2 + (-1)**(1-z0x1) * acc.z0x1 + ## w0z1x0 <- (1-z1x0)**2 + (-1)**(1-z1x0) * acc.z1x0 + ## w0z1x1 <- (1-z1x1)**2 + (-1)**(1-z1x1) * acc.z1x1 + + ## ##perrorz0 <- w0z0*(pyz0) + ## ##perrorz1 <- w0z1*(pyz1) + + ## w0z0x0.noisy.odds <- rlogis(nz0x0,qlogis(w0z0x0)) + ## w0z0x1.noisy.odds <- rlogis(nz0x1,qlogis(w0z0x1)) + ## w0z1x0.noisy.odds <- rlogis(nz1x0,qlogis(w0z1x0)) + ## w0z1x1.noisy.odds <- rlogis(nz1x1,qlogis(w0z1x1)) + + ## df[(z==0)&(x==0),w:=plogis(w0z0x0.noisy.odds)] + ## df[(z==0)&(x==1),w:=plogis(w0z0x1.noisy.odds)] + ## df[(z==1)&(x==0),w:=plogis(w0z1x0.noisy.odds)] + ## df[(z==1)&(x==1),w:=plogis(w0z1x1.noisy.odds)] + + ## df[,w_pred:=as.integer(w > 0.5)] + ## print(mean(df[z==0]$x == df[z==0]$w_pred)) + ## print(mean(df[z==1]$x == df[z==1]$w_pred)) + ## print(mean(df$w_pred == df$x)) + + + + resids <- resid(lm(y~x + z)) + odds.x1 <- qlogis(prediction_accuracy) + y_bias*qlogis(pnorm(resids[x==1],log.p=T),log.p=T) + z_bias * qlogis(pnorm(z[x==1],sd(z),log.p=T),log.p=T) + odds.x0 <- qlogis(prediction_accuracy,lower.tail=F) + y_bias*qlogis(pnorm(resids[x==0],log.p=T),log.p=T) + z_bias * qlogis(pnorm(z[x==0],sd(z),log.p=T),log.p=T) + + ## acc.x0 <- p.correct[df[,x==0]] + ## acc.x1 <- p.correct[df[,x==1]] + + df[x==0,w:=plogis(rlogis(.N,odds.x0))] + df[x==1,w:=plogis(rlogis(.N,odds.x1))] + + print(prediction_accuracy) + print(resids[is.na(df$w)]) + print(odds.x0[is.na(df$w)]) + print(odds.x1[is.na(df$w)]) + + df[,w_pred := as.integer(w > 0.5)] + + + print(mean(df$w_pred == df$x)) + print(mean(df[y>=0]$w_pred == df[y>=0]$x)) + print(mean(df[y<=0]$w_pred == df[y<=0]$x)) + return(df) +} + +parser <- arg_parser("Simulate data and fit corrected models") +parser <- add_argument(parser, "--N", default=5000, help="number of observations of w") +parser <- add_argument(parser, "--m", default=500, help="m the number of ground truth observations") +parser <- add_argument(parser, "--seed", default=51, help='seed for the rng') +parser <- add_argument(parser, "--outfile", help='output file', default='example_2.feather') +parser <- add_argument(parser, "--y_explained_variance", help='what proportion of the variance of y can be explained?', default=0.1) +parser <- add_argument(parser, "--prediction_accuracy", help='how accurate is the predictive model?', default=0.75) +parser <- add_argument(parser, "--accuracy_imbalance_difference", help='how much more accurate is the predictive model for one class than the other?', default=0.3) +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 z on y', default=0.3) +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*x") +parser <- add_argument(parser, "--y_bias", help='coefficient of y on the probability a classification is correct', default=-0.5) +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') +parser <- add_argument(parser, "--sd_y_mixin", help='varience of the non-normal part of Y', default=10) +args <- parse_args(parser) + +B0 <- 0 +Px <- args$Px +Bxy <- args$Bxy +Bzy <- args$Bzy +Bzx <- args$Bzx + +if(args$m < args$N){ + + df <- simulate_data(args$N, args$m, B0, Bxy, Bzx, Bzy, args$seed, args$y_explained_variance, args$prediction_accuracy, y_bias=args$y_bias, sd_y_mixin=args$sd_y_mixin) + + ## df.pc <- df[,.(x,y,z,w_pred,w)] + ## # df.pc <- df.pc[,err:=x-w_pred] + ## 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$truth_formula, confint_method=args$confint_method, error='', 'sd_y_mixin'=args$sd_y_mixin) + + outline <- run_simulation(df, result, outcome_formula=as.formula(args$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$outfile)){ + logdata <- read_feather(args$outfile) + logdata <- rbind(logdata,as.data.table(outline), fill=TRUE) + } else { + logdata <- as.data.table(outline) + } + + print(outline) + write_feather(logdata, args$outfile) + unlock(outfile_lock) +} diff --git a/simulations/Makefile b/simulations/Makefile index 821280b..c4d60b8 100644 --- a/simulations/Makefile +++ b/simulations/Makefile @@ -8,7 +8,7 @@ explained_variances=[0.1] all:main supplement main:remembr.RDS -supplement:robustness_1.RDS robustness_1_dv.RDS robustness_2.RDS robustness_2_dv.RDS robustness_3.RDS robustness_3_dv.RDS robustness_3_proflik.RDS robustness_3_dv_proflik.RDS robustness_4.RDS robustness_4_dv.RDS +supplement:robustness_1.RDS robustness_1_dv.RDS robustness_2.RDS robustness_2_dv.RDS robustness_3.RDS robustness_3_dv.RDS robustness_3_proflik.RDS robustness_3_dv_proflik.RDS robustness_4.RDS robustness_4_dv.RDS robustness_5.RDS robustness_5_dv.RDS robustness_6.feather srun=sbatch --wait --verbose run_job.sbatch @@ -16,7 +16,7 @@ srun=sbatch --wait --verbose run_job.sbatch joblists:example_1_jobs example_2_jobs example_3_jobs # test_true_z_jobs: test_true_z.R simulation_base.R -# sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript test_true_z.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["test_true_z.feather"], "y_explained_variancevari":${explained_variances}, "Bzx":${Bzx}}' --outfile test_true_z_jobsb +# sbatch --wait --verbose run_job.sbatch ./grid_sweep.py --command "Rscript test_true_z.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["test_true_z.feather"], "y_explained_variancevari":${explained_variances}, "Bzx":${Bzx}}' --outfile test_true_z_jobsb # test_true_z.feather: test_true_z_jobs # rm -f test_true_z.feather @@ -25,48 +25,46 @@ joblists:example_1_jobs example_2_jobs example_3_jobs example_1_jobs: 01_two_covariates.R simulation_base.R grid_sweep.py pl_methods.R - ${srun} grid_sweep.py --command "Rscript 01_two_covariates.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_1.feather"], "y_explained_variance":${explained_variances}, "Bzx":[1]}' --outfile example_1_jobs + ${srun} ./grid_sweep.py --command "Rscript 01_two_covariates.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_1.feather"], "y_explained_variance":${explained_variances}, "Bzx":[1]}' --outfile example_1_jobs example_1.feather: example_1_jobs rm -f example_1.feather sbatch --wait --verbose --array=1-1000 run_simulation.sbatch 0 example_1_jobs sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_1_jobs sbatch --wait --verbose --array=2001-3000 run_simulation.sbatch 0 example_1_jobs - sbatch --wait --verbose --array=3001-4000 run_simulation.sbatch 0 example_1_jobs - sbatch --wait --verbose --array=4001-$(shell cat example_1_jobs | wc -l) run_simulation.sbatch 0 example_1_jobs + sbatch --wait --verbose --array=3001-$(shell cat example_1_jobs | wc -l) run_simulation.sbatch 0 example_1_jobs example_2_jobs: 02_indep_differential.R simulation_base.R grid_sweep.py pl_methods.R - sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+z+x"]}' --outfile example_2_jobs + sbatch --wait --verbose run_job.sbatch ./grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+z+x"]}' --outfile example_2_jobs example_2.feather: example_2_jobs rm -f example_2.feather sbatch --wait --verbose --array=1-1000 run_simulation.sbatch 0 example_2_jobs sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_2_jobs sbatch --wait --verbose --array=2001-3000 run_simulation.sbatch 0 example_2_jobs - sbatch --wait --verbose --array=3001-4000 run_simulation.sbatch 0 example_2_jobs - sbatch --wait --verbose --array=4001-$(shell cat example_2_jobs | wc -l) + sbatch --wait --verbose --array=3001-$(shell cat example_2_jobs | wc -l) run_simulation.sbatch 0 example_2_jobs + # example_2_B_jobs: example_2_B.R -# sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript example_2_B.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2_B.feather"]}' --outfile example_2_B_jobs +# sbatch --wait --verbose run_job.sbatch python3 ./grid_sweep.py --command "Rscript example_2_B.R" --arg_dict '{"N":${Ns},"m":${ms}, "seed":${seeds}, "outfile":["example_2_B.feather"]}' --outfile example_2_B_jobs # example_2_B.feather: example_2_B_jobs # rm -f example_2_B.feather # sbatch --wait --verbose --array=1-3000 run_simulation.sbatch 0 example_2_B_jobs example_3_jobs: 03_depvar.R simulation_base.R grid_sweep.py pl_methods.R - sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 03_depvar.R" --arg_dict '{"N":${Ns},"m":${ms}, "Bxy":[0.7],"Bzy":[-0.7],"Bzx":[1],"seed":${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":${explained_variances}}' --outfile example_3_jobs example_3.feather: example_3_jobs rm -f example_3.feather sbatch --wait --verbose --array=1-1000 run_simulation.sbatch 0 example_3_jobs sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_3_jobs sbatch --wait --verbose --array=2001-3000 run_simulation.sbatch 0 example_3_jobs - sbatch --wait --verbose --array=3001-4000 run_simulation.sbatch 0 example_3_jobs - sbatch --wait --verbose --array=4001-$(shell cat example_3_jobs | wc -l) run_simulation.sbatch 0 example_3_jobs + sbatch --wait --verbose --array=3001-$(shell cat example_3_jobs | wc -l) run_simulation.sbatch 0 example_3_jobs example_4_jobs: 04_depvar_differential.R simulation_base.R grid_sweep.py pl_methods.R - sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 04_depvar_differential.R" --arg_dict '{"N":${Ns},"Bxy":[0.7],"Bzy":[-0.7],"Bzx":[1], "m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[-0.5], "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":${seeds}, "outfile":["example_4.feather"], "z_bias":[-0.5], "prediction_accuracy":[0.73]}' --outfile example_4_jobs example_4.feather: example_4_jobs rm -f example_4.feather @@ -74,9 +72,7 @@ example_4.feather: example_4_jobs sbatch --wait --verbose --array=1001-2000 run_simulation.sbatch 0 example_4_jobs sbatch --wait --verbose --array=2001-3001 run_simulation.sbatch 0 example_4_jobs sbatch --wait --verbose --array=2001-3000 run_simulation.sbatch 0 example_4_jobs - sbatch --wait --verbose --array=3001-4000 run_simulation.sbatch 0 example_4_jobs - sbatch --wait --verbose --array=4001-$(shell cat example_4_jobs | wc -l) run_simulation.sbatch 0 example_4_jobs - + sbatch --wait --verbose --array=3001-$(shell cat example_4_jobs | wc -l) run_simulation.sbatch 0 example_4_jobs remembr.RDS:example_1.feather example_2.feather example_3.feather example_4.feather plot_example.R plot_dv_example.R summarize_estimator.R @@ -92,22 +88,21 @@ STEP=1000 ONE=1 robustness_Ns=[1000,5000] -robustness_robustness_ms=[100,200] +robustness_ms=[100,200] #in robustness 1 / example 2 misclassification is correlated with Y. robustness_1_jobs_p1: 02_indep_differential.R simulation_base.R grid_sweep.py - sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":[1000],"m":${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 + 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":${explained_variances}, "Bzy":[-0.3,0],"Bxy":[0.3],"Bzx":[1,0], "outcome_formula":["y~x+z"], "z_bias":[0, 0.5], "proxy_formula":["w_pred~y+x"], "truth_formula":["x~1"]}' --outfile robustness_1_jobs_p1 robustness_1_jobs_p2: 02_indep_differential.R simulation_base.R grid_sweep.py - sbatch --wait --verbose run_job.sbatch grid_sweep.py --command "Rscript 02_indep_differential.R" --arg_dict '{"N":[5000],"m":${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 + 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.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 ITEMS_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)) - + $(eval ITEMS_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;) @@ -117,10 +112,10 @@ robustness_1.RDS: robustness_1.feather summarize_estimator.R # 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 + ${srun} ./grid_sweep.py --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":[1000],"Bzx":[1], "Bxy":[0.7,0],"Bzy":[-0.7,0],"m":${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_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 + ${srun} ./grid_sweep.py --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":[5000],"Bzx":[1], "Bxy":[0.7,0],"Bzy":[-0.7,0],"m":${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 $@ @@ -136,21 +131,21 @@ robustness_1_dv.RDS: robustness_1_dv.feather summarize_estimator.R ${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 +robustness_2_jobs_p1: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_jobs_p2: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_jobs_p3: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_jobs_p4: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 $@ @@ -172,21 +167,21 @@ 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 +robustness_2_dv_jobs_p1: grid_sweep.py 03_depvar.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_dv_jobs_p2: grid_sweep.py 03_depvar.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_dv_jobs_p3: grid_sweep.py 03_depvar.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_2_dv_jobs_p4: grid_sweep.py 03_depvar.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 $@ @@ -209,9 +204,9 @@ robustness_2_dv.RDS: plot_dv_example.R robustness_2_dv.feather summarize_estimat ${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 +robustness_3_proflik_jobs: grid_sweep.py 01_two_covariates.R simulation_base.R 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 $@ + ${srun} ./$< --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 $@ @@ -224,17 +219,17 @@ robustness_3_proflik.RDS: plot_example.R robustness_3_proflik.feather summarize_ ${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 +robustness_3_jobs_p1: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_3_jobs_p2: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 +robustness_3_jobs_p3: grid_sweep.py 01_two_covariates.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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 $@ @@ -253,9 +248,9 @@ 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_proflik_jobs: 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 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 $@ + ${srun} ./$< --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 $@ @@ -268,20 +263,18 @@ robustness_3_dv_proflik.RDS: plot_dv_example.R robustness_3_dv_proflik.feather s ${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 + robustness_3_dv_jobs_p1: grid_sweep.py 03_depvar.R simulation_base.R 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 $@ + ${srun} ./$< --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 +robustness_3_dv_jobs_p2: grid_sweep.py 03_depvar.R simulation_base.R 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":[0.847,1.386], "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 +robustness_3_dv_jobs_p3: grid_sweep.py 03_depvar.R simulation_base.R 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":[2.197,2.944], "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 $@ @@ -303,17 +296,22 @@ robustness_3_dv.RDS: plot_dv_example.R robustness_3_dv.feather summarize_estimat -robustness_4_jobs_p1: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py +robustness_4_jobs_p1: grid_sweep.py 02_indep_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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+z"], "truth_formula":["x~z"], "y_bias":[-2.944,-2.197]}' --outfile $@ -robustness_4_jobs_p2: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py +robustness_4_jobs_p2: grid_sweep.py 02_indep_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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+z"], "truth_formula":["x~z"], "y_bias":[-1.386,-0.846]}' --outfile $@ + +robustness_4_jobs_p3: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f ./$@ + ${srun} ./$< --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+z"], "truth_formula":["x~z"], "y_bias":[-0.405,-0.25]}' --outfile $@ + +robustness_4_jobs_p4: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f ./$@ + ${srun} ./$< --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+z"], "truth_formula":["x~z"], "y_bias":[0,-0.1]}' --outfile $@ -robustness_4_jobs_p3: grid_sweep.py 02_indep_differential.R simulation_base.R grid_sweep.py - rm -f $@ - ${srun} $< --command 'Rscript 02_indep_differential.R' --arg_dict '{"N":${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 $@ @@ -323,10 +321,15 @@ robustness_4.feather: robustness_4_jobs_p1 robustness_4_jobs_p2 robustness_4_job $(eval ITEMS_2!=seq $(START) $(STEP) $(END_2)) $(eval END_3!=cat robustness_4_jobs_p3 | wc -l) $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3)) + $(eval END_4!=cat robustness_4_jobs_p3 | wc -l) + $(eval ITEMS_3!=seq $(START) $(STEP) $(END_4)) + $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_4_jobs_p1;) $(foreach item,$(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;) + $(foreach item,$(ITEMS_4),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 summarize_estimator.R rm -f $@ @@ -335,34 +338,32 @@ robustness_4.RDS: plot_example.R robustness_4.feather summarize_estimator.R # '{"N":${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 +robustness_4_dv_jobs_p1: grid_sweep.py 04_depvar_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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"], "z_bias":[0,0.1]}' --outfile $@ -robustness_4_dv_jobs_p2: grid_sweep.py 04_depvar_differential.R simulation_base.R grid_sweep.py +robustness_4_dv_jobs_p2: grid_sweep.py 04_depvar_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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"], "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 +robustness_4_dv_jobs_p3: grid_sweep.py 04_depvar_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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"],"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 +robustness_4_dv_jobs_p4: grid_sweep.py 04_depvar_differential.R simulation_base.R rm -f $@ - ${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 $@ + ${srun} ./$< --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"], "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_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 END_2!=cat robustness_4_dv_jobs_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 END_3!=cat robustness_4_dv_jobs_p3 | wc -l) $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3)) - - + $(eval END_4!=cat robustness_4_dv_jobs_p4 | wc -l) + $(eval ITEMS_4!=seq $(START) $(STEP) $(END_4)) $(foreach item,$(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;) @@ -371,7 +372,86 @@ robustness_4_dv.feather: robustness_4_dv_jobs_p1 robustness_4_dv_jobs_p2 robustn 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 $@ + ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_4_dv" --remember-file $@ + + +robustness_5_jobs_p1: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 02_indep_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[1.386,2.197], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"]}' --outfile $@ + +robustness_5_jobs_p2: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 02_indep_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0.405,0.846], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"]}' --outfile $@ + +robustness_5_jobs_p3: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 02_indep_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[0,0.25], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"]}' --outfile $@ + +robustness_5_jobs_p4: grid_sweep.py 02_indep_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 02_indep_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3],"Bzx":[2.944], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"]}' --outfile $@ + + +robustness_5.feather: robustness_5_jobs_p1 robustness_5_jobs_p2 robustness_5_jobs_p3 + rm -f $@ + $(eval END_1!=cat robustness_5_jobs_p1 | wc -l) + $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1)) + $(eval END_2!=cat robustness_5_jobs_p2 | wc -l) + $(eval ITEMS_2!=seq $(START) $(STEP) $(END_2)) + $(eval END_3!=cat robustness_5_jobs_p3 | wc -l) + $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3)) + $(eval END_4!=cat robustness_5_jobs_p4 | wc -l) + $(eval ITEMS_4!=seq $(START) $(STEP) $(END_4)) + + $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_jobs_p1;) + $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_jobs_p2;) + $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_jobs_p3;) + $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_jobs_p4;) + +robustness_5.RDS: plot_example.R robustness_5.feather summarize_estimator.R + rm -f $@ + ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_5" --remember-file $@ + + +# '{"N":${robustness_Ns},"Bxy":[0.7],"Bzy":[-0.7],"m":${ms}, "seed":${seeds}, "outfile":["example_4.feather"], "z_bias":[0.5]}' --example_4_jobs + +robustness_5_dv_jobs_p1: grid_sweep.py 04_depvar_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[0,0.25], "outcome_formula":["y~x+z"], "z_bias":[-0.5]}' --outfile $@ + +robustness_5_dv_jobs_p2: grid_sweep.py 04_depvar_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[0.405,0.846], "outcome_formula":["y~x+z"], "z_bias":[-0.5]}' --outfile $@ + +robustness_5_dv_jobs_p3: grid_sweep.py 04_depvar_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5_dv.feather"], "Bzy":[-0.7],"Bxy":[0.7],"Bzx":[1.386,2.197],"outcome_formula":["y~x+z"], "z_bias":[-0.5]}' --outfile $@ + +robustness_5_dv_jobs_p4: grid_sweep.py 04_depvar_differential.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 04_depvar_differential.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_5_dv.feather"],"Bzy":[-0.7],"Bxy":[0.7],"Bzx":[2.944], "outcome_formula":["y~x+z"], "z_bias":[-0.5]}' --outfile $@ + +robustness_5_dv.feather: robustness_5_dv_jobs_p1 robustness_5_dv_jobs_p2 robustness_5_dv_jobs_p3 robustness_5_dv_jobs_p4 + rm -f $@ + $(eval END_1!=cat robustness_5_dv_jobs_p1 | wc -l) + $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1)) + $(eval END_2!=cat robustness_5_dv_jobs_p2 | wc -l) + $(eval ITEMS_2!=seq $(START) $(STEP) $(END_2)) + $(eval END_3!=cat robustness_5_dv_jobs_p3 | wc -l) + $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3)) + $(eval END_4!=cat robustness_5_dv_jobs_p4 | wc -l) + $(eval ITEMS_4!=seq $(START) $(STEP) $(END_4)) + + + $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_dv_jobs_p1;) + $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_dv_jobs_p2;) + $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_dv_jobs_p3;) + $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_5_dv_jobs_p4;) + + +robustness_5_dv.RDS: plot_dv_example.R robustness_5_dv.feather summarize_estimator.R + rm -f $@ + ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_5_dv" --remember-file $@ clean_main: @@ -404,5 +484,44 @@ clean_all: # sbatch --wait --verbose --array=1-3000 run_simulation.sbatch 0 example_2_B_mecor_jobs # sbatch --wait --verbose --array=3001-6001 run_simulation.sbatch 0 example_2_B_mecor_jobs +robustness_6_jobs_p1: grid_sweep.py 03_indep_differential_nonnorm.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 03_indep_differential_nonnorm.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_6.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"],"sd_y_mixin":[0,1,2.5]}' --outfile $@ + +robustness_6_jobs_p2: grid_sweep.py 03_indep_differential_nonnorm.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 03_indep_differential_nonnorm.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_6.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"],"sd_y_mixin":[5,10]}' --outfile $@ + +robustness_6_jobs_p3: grid_sweep.py 03_indep_differential_nonnorm.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 03_indep_differential_nonnorm.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_6.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"],"sd_y_mixin":[0,1,2.5],"y_bias":[0]}' --outfile $@ + +robustness_6_jobs_p4: grid_sweep.py 03_indep_differential_nonnorm.R simulation_base.R + rm -f $@ + ${srun} ./$< --command 'Rscript 03_indep_differential_nonnorm.R' --arg_dict '{"N":${robustness_Ns},"m":${robustness_ms}, "seed":${seeds}, "outfile":["robustness_6.feather"],"y_explained_variance":${explained_variances}, "Bzy":[-0.3],"Bxy":[0.3], "outcome_formula":["y~x+z"], "proxy_formula":["w_pred~y+x+z"], "truth_formula":["x~z"],"sd_y_mixin":[5,10],"y_bias":[0]}' --outfile $@ + + +robustness_6.feather: robustness_6_jobs_p1 robustness_6_jobs_p2 robustness_6_jobs_p3 robustness_6_jobs_p4 + rm -f $@ + $(eval END_1!=cat robustness_6_jobs_p1 | wc -l) + $(eval ITEMS_1!=seq $(START) $(STEP) $(END_1)) + $(eval END_2!=cat robustness_6_jobs_p2 | wc -l) + $(eval ITEMS_2!=seq $(START) $(STEP) $(END_2)) + $(eval END_3!=cat robustness_6_jobs_p3 | wc -l) + $(eval ITEMS_3!=seq $(START) $(STEP) $(END_3)) + $(eval END_4!=cat robustness_6_jobs_p4 | wc -l) + $(eval ITEMS_4!=seq $(START) $(STEP) $(END_4)) + + + # $(foreach item,$(ITEMS_1),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_6_jobs_p1;) + # $(foreach item,$(ITEMS_2),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_6_jobs_p2;) + # $(foreach item,$(ITEMS_3),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_6_jobs_p3;) + $(foreach item,$(ITEMS_4),sbatch --wait --verbose --array=$(shell expr $(item) + $(ONE))-$(shell expr $(item) + $(STEP)) run_simulation.sbatch 0 robustness_6_jobs_p4;) + + +robustness_6.RDS: plot_example.R robustness_6.feather summarize_estimator.R + rm -f $@ + ${srun} Rscript $< --infile $(word 2, $^) --name "robustness_6" --remember-file $@ + .PHONY: supplement diff --git a/simulations/measerr_methods.R b/simulations/measerr_methods.R index 98ab28d..8f8eb5d 100644 --- a/simulations/measerr_methods.R +++ b/simulations/measerr_methods.R @@ -23,7 +23,7 @@ likelihood.logistic <- function(model.params, outcome, model.matrix){ } ## outcome_formula <- y ~ x + z; proxy_formula <- w_pred ~ y + x + z + x:z + x:y + z:y -measerr_mle_dv <- function(df, outcome_formula, outcome_family=binomial(link='logit'), proxy_formula, proxy_family=binomial(link='logit'),maxit=1e6, method='optim'){ +measerr_mle_dv <- function(df, outcome_formula, outcome_family=binomial(link='logit'), proxy_formula, proxy_family=binomial(link='logit'),maxit=1e6, method='optim',optim_method='L-BFGS-B'){ df.obs <- model.frame(outcome_formula, df) proxy.model.matrix <- model.matrix(proxy_formula, df) proxy.variable <- all.vars(proxy_formula)[1] diff --git a/simulations/pl_methods.R b/simulations/pl_methods.R index f014eec..2099f1a 100644 --- a/simulations/pl_methods.R +++ b/simulations/pl_methods.R @@ -51,19 +51,20 @@ zhang.mle.iv <- function(df){ fn <- df.obs[(w_pred==0) & (x.obs==1), .N] npv <- tn / (tn + fn) + tp <- df.obs[(w_pred==1) & (x.obs == w_pred),.N] fp <- df.obs[(w_pred==1) & (x.obs == 0),.N] ppv <- tp / (tp + fp) - nll <- function(B0=0, Bxy=0, Bzy=0, sigma_y=0.1){ + nll <- function(B0=0, Bxy=0, Bzy=0, sigma_y=9){ ## fpr = 1 - TNR ### Problem: accounting for uncertainty in ppv / npv ## fnr = 1 - TPR ll.y.obs <- with(df.obs, dnorm(y, B0 + Bxy * x + Bzy * z, sd=sigma_y,log=T)) + ll <- sum(ll.y.obs) - # unobserved case; integrate out x ll.x.1 <- with(df.unobs, dnorm(y, B0 + Bxy + Bzy * z, sd = sigma_y, log=T)) ll.x.0 <- with(df.unobs, dnorm(y, B0 + Bzy * z, sd = sigma_y,log=T)) @@ -75,10 +76,11 @@ zhang.mle.iv <- function(df){ lls.x.0 <- colLogSumExps(rbind(log(1-npv) + ll.x.1, log(npv) + ll.x.0)) lls <- colLogSumExps(rbind(df.unobs$w_pred * lls.x.1, (1-df.unobs$w_pred) * lls.x.0)) + ll <- ll + sum(lls) - return(-ll) + } - mlefit <- mle2(minuslogl = nll, control=list(maxit=1e6), lower=list(sigma_y=0.0001, B0=-Inf, Bxy=-Inf, Bzy=-Inf), + mlefit <- mle2(minuslogl = nll, control=list(maxit=1e6), lower=list(sigma_y=0.00001, B0=-Inf, Bxy=-Inf, Bzy=-Inf), upper=list(sigma_y=Inf, B0=Inf, Bxy=Inf, Bzy=Inf),method='L-BFGS-B') return(mlefit) } diff --git a/simulations/simulation_base.R b/simulations/simulation_base.R index bafd7d3..af03408 100644 --- a/simulations/simulation_base.R +++ b/simulations/simulation_base.R @@ -151,21 +151,11 @@ run_simulation_depvar <- function(df, result, outcome_formula=y~x+z, proxy_formu temp.df <- copy(df) temp.df[,y:=y.obs] - if(confint_method=='quad'){ - mod.caroll.lik <- measerr_mle_dv(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula) - fischer.info <- solve(mod.caroll.lik$hessian) - coef <- mod.caroll.lik$par - ci.upper <- coef + sqrt(diag(fischer.info)) * 1.96 - ci.lower <- coef - sqrt(diag(fischer.info)) * 1.96 - } - else{ ## confint_method is 'profile' - - mod.caroll.lik <- measerr_mle_dv(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, method='bbmle') - coef <- coef(mod.caroll.lik) - ci <- confint(mod.caroll.lik, method='spline') - ci.lower <- ci[,'2.5 %'] - ci.upper <- ci[,'97.5 %'] - } + mod.caroll.lik <- measerr_mle_dv(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula) + fischer.info <- solve(mod.caroll.lik$hessian) + coef <- mod.caroll.lik$par + ci.upper <- coef + sqrt(diag(fischer.info)) * 1.96 + ci.lower <- coef - sqrt(diag(fischer.info)) * 1.96 result <- append(result, list(Bxy.est.mle = coef['x'], @@ -175,6 +165,19 @@ run_simulation_depvar <- function(df, result, outcome_formula=y~x+z, proxy_formu Bzy.ci.upper.mle = ci.upper['z'], Bzy.ci.lower.mle = ci.lower['z'])) + mod.caroll.profile.lik <- measerr_mle_dv(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, method='bbmle') + coef <- coef(mod.caroll.profile.lik) + ci <- confint(mod.caroll.profile.lik, method='spline') + ci.lower <- ci[,'2.5 %'] + ci.upper <- ci[,'97.5 %'] + + result <- append(result, + list(Bxy.est.mle.profile = coef['x'], + Bxy.ci.upper.mle.profile = ci.upper['x'], + Bxy.ci.lower.mle.profile = ci.lower['x'], + Bzy.est.mle.profile = coef['z'], + Bzy.ci.upper.mle.profile = ci.upper['z'], + Bzy.ci.lower.mle.profile = ci.lower['z'])) ## my implementatoin of liklihood based correction mod.zhang <- zhang.mle.dv(df) @@ -201,8 +204,8 @@ run_simulation_depvar <- function(df, result, outcome_formula=y~x+z, proxy_formu ) tryCatch({ - amelia.out.k <- amelia(df, m=200, p2s=0, idvars=c('y','ystar','w')) - mod.amelia.k <- zelig(y.obs~x+z, model='ls', data=amelia.out.k$imputations, cite=FALSE) + amelia.out.k <- amelia(df, m=200, p2s=0, idvars=c('y','ystar','w'),ords="y.obs") + mod.amelia.k <- zelig(y.obs~x+z, model='logit', data=amelia.out.k$imputations, cite=FALSE) (coefse <- combine_coef_se(mod.amelia.k, messages=FALSE)) est.x.mi <- coefse['x','Estimate'] est.x.se <- coefse['x','Std.Error'] @@ -340,44 +343,72 @@ run_simulation <- function(df, result, outcome_formula=y~x+z, proxy_formula=NUL tryCatch({ temp.df <- copy(df) temp.df <- temp.df[,x:=x.obs] - if(confint_method=='quad'){ - mod.caroll.lik <- measerr_mle(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, truth_formula=truth_formula, method='optim') - fischer.info <- solve(mod.caroll.lik$hessian) - coef <- mod.caroll.lik$par - ci.upper <- coef + sqrt(diag(fischer.info)) * 1.96 - ci.lower <- coef - sqrt(diag(fischer.info)) * 1.96 - } else { # confint_method == 'bbmle' - - mod.caroll.lik <- measerr_mle(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, truth_formula=truth_formula, method='bbmle') - coef <- coef(mod.caroll.lik) - ci <- confint(mod.caroll.lik, method='spline') - ci.lower <- ci[,'2.5 %'] - ci.upper <- ci[,'97.5 %'] - } + mod.caroll.lik <- measerr_mle(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, truth_formula=truth_formula, method='optim') + fischer.info <- solve(mod.caroll.lik$hessian) + coef <- mod.caroll.lik$par + ci.upper <- coef + sqrt(diag(fischer.info)) * 1.96 + ci.lower <- coef - sqrt(diag(fischer.info)) * 1.96 + mle_result <- list(Bxy.est.mle = coef['x'], Bxy.ci.upper.mle = ci.upper['x'], Bxy.ci.lower.mle = ci.lower['x'], Bzy.est.mle = coef['z'], Bzy.ci.upper.mle = ci.upper['z'], Bzy.ci.lower.mle = ci.lower['z']) + }, + error=function(e) {result[['error']] <- as.character(e) + }) + + result <- append(result, mle_result) + mle_result_proflik <- list(Bxy.est.mle.profile = NULL, + Bxy.ci.upper.mle.profile = NULL, + Bxy.ci.lower.mle.profile = NULL, + Bzy.est.mle.profile = NULL, + Bzy.ci.upper.mle.profile = NULL, + Bzy.ci.lower.mle.profile = NULL) + + tryCatch({ + ## confint_method == 'bbmle' + mod.caroll.lik <- measerr_mle(temp.df, outcome_formula=outcome_formula, proxy_formula=proxy_formula, truth_formula=truth_formula, method='bbmle') + coef <- coef(mod.caroll.lik) + ci <- confint(mod.caroll.lik, method='spline') + ci.lower <- ci[,'2.5 %'] + ci.upper <- ci[,'97.5 %'] + + mle_result_proflik <- list(Bxy.est.mle.profile = coef['x'], + Bxy.ci.upper.mle.profile = ci.upper['x'], + Bxy.ci.lower.mle.profile = ci.lower['x'], + Bzy.est.mle.profile = coef['z'], + Bzy.ci.upper.mle.profile = ci.upper['z'], + Bzy.ci.lower.mle.profile = ci.lower['z']) }, error=function(e) {result[['error']] <- as.character(e) }) - - result <- append(result, mle_result) + result <- append(result, mle_result_proflik) + zhang_result <- list(Bxy.est.mle.zhang = NULL, + Bxy.ci.upper.mle.zhang = NULL, + Bxy.ci.lower.mle.zhang = NULL, + Bzy.est.mle.zhang = NULL, + Bzy.ci.upper.mle.zhang = NULL, + Bzy.ci.lower.mle.zhang = NULL) + + tryCatch({ mod.zhang.lik <- zhang.mle.iv(df) coef <- coef(mod.zhang.lik) ci <- confint(mod.zhang.lik,method='quad') - result <- append(result, - list(Bxy.est.zhang = coef['Bxy'], - Bxy.ci.upper.zhang = ci['Bxy','97.5 %'], - Bxy.ci.lower.zhang = ci['Bxy','2.5 %'], - Bzy.est.zhang = coef['Bzy'], - Bzy.ci.upper.zhang = ci['Bzy','97.5 %'], - Bzy.ci.lower.zhang = ci['Bzy','2.5 %'])) + zhang_result <- list(Bxy.est.zhang = coef['Bxy'], + Bxy.ci.upper.zhang = ci['Bxy','97.5 %'], + Bxy.ci.lower.zhang = ci['Bxy','2.5 %'], + Bzy.est.zhang = coef['Bzy'], + Bzy.ci.upper.zhang = ci['Bzy','97.5 %'], + Bzy.ci.lower.zhang = ci['Bzy','2.5 %']) + }, + error=function(e) {result[['error']] <- as.character(e) + }) + result <- append(result, zhang_result) ## What if we can't observe k -- most realistic scenario. We can't include all the ML features in a model. ## amelia.out.nok <- amelia(df, m=200, p2s=0, idvars=c("x","w_pred"), noms=noms) diff --git a/simulations/summarize_estimator.R b/simulations/summarize_estimator.R index 1e1341d..8ddeb7c 100644 --- a/simulations/summarize_estimator.R +++ b/simulations/summarize_estimator.R @@ -1,5 +1,6 @@ +library(ggdist) -summarize.estimator <- function(df, suffix='naive', coefname='x'){ +summarize.estimator <- function(sims.df, suffix='naive', coefname='x'){ reported_vars <- c( 'Bxy', @@ -13,10 +14,10 @@ summarize.estimator <- function(df, suffix='naive', coefname='x'){ grouping_vars <- grouping_vars[grouping_vars %in% names(df)] - part <- df[, - c(reported_vars, - grouping_vars), - with=FALSE] + part <- sims.df[, + unique(c(reported_vars, + grouping_vars)), + with=FALSE] true.in.ci <- as.integer((part$Bxy >= part[[paste0('B',coefname,'y.ci.lower.',suffix)]]) & (part$Bxy <= part[[paste0('B',coefname,'y.ci.upper.',suffix)]])) @@ -29,6 +30,7 @@ summarize.estimator <- function(df, suffix='naive', coefname='x'){ bias=bias, sign.correct =sign.correct)] + part.plot <- part[, .(p.true.in.ci = mean(true.in.ci), mean.bias = mean(bias), mean.est = mean(.SD[[paste0('B',coefname,'y.est.',suffix)]],na.rm=T),