X-Git-Url: https://code.communitydata.science/ml_measurement_error_public.git/blobdiff_plain/5c931a7198452ff3ce0ace5b1f68046bfb33d352..214551f74cc94ef1fa2f24aa317265c25bb03757:/simulations/04_depvar_differential.R diff --git a/simulations/04_depvar_differential.R b/simulations/04_depvar_differential.R index df0e825..2c43f59 100644 --- a/simulations/04_depvar_differential.R +++ b/simulations/04_depvar_differential.R @@ -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)) @@ -70,29 +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, "--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 <- 0 +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)