From: Nathan TeBlunthuis Date: Fri, 12 Aug 2022 00:34:56 +0000 (-0700) Subject: simplify simulation 02. X-Git-Url: https://code.communitydata.science/ml_measurement_error_public.git/commitdiff_plain/2cd447c327744263d5f94b20e1146cdf31b2ec2c?hp=27a3849d5dcddb668970b283f6aac0d7c8fb642e simplify simulation 02. --- diff --git a/simulations/02_indep_differential.R b/simulations/02_indep_differential.R index 7e2e428..cee3643 100644 --- a/simulations/02_indep_differential.R +++ b/simulations/02_indep_differential.R @@ -31,7 +31,7 @@ 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, Bzx, Bzy, seed, y_explained_variance=0.025, prediction_accuracy=0.73, accuracy_imbalance_difference=0.3){ +simulate_data <- function(N, m, B0, Bxy, Bzx, Bzy, seed, y_explained_variance=0.025, prediction_accuracy=0.73, y_bias=-0.8){ set.seed(seed) # make w and y dependent z <- rbinom(N, 1, 0.5) @@ -49,107 +49,59 @@ simulate_data <- function(N, m, B0, Bxy, Bzx, Bzy, seed, y_explained_variance=0. df <- df[, x.obs := x] } - ## px <- mean(x) - ## accuracy_imbalance_ratio <- (prediction_accuracy + accuracy_imbalance_difference/2) / (prediction_accuracy - accuracy_imbalance_difference/2) - - ## # this works because of conditional probability - ## accuracy_x0 <- prediction_accuracy / (px*(accuracy_imbalance_ratio) + (1-px)) - ## accuracy_x1 <- accuracy_imbalance_ratio * accuracy_x0 - - ## x0 <- df[x==0]$x - ## x1 <- df[x==1]$x - ## nx1 <- nrow(df[x==1]) - ## nx0 <- nrow(df[x==0]) - - ## yx0 <- df[x==0]$y - ## yx1 <- df[x==1]$y - - # tranform yz0.1 into a logistic distribution with mean accuracy_z0 - ## acc.x0 <- plogis(0.5*scale(yx0) + qlogis(accuracy_x0)) - ## acc.x1 <- plogis(1.5*scale(yx1) + qlogis(accuracy_x1)) - - ## w0x0 <- (1-x0)**2 + (-1)**(1-x0) * acc.x0 - ## w0x1 <- (1-x1)**2 + (-1)**(1-x1) * acc.x1 - 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)] + ## probablity of an error is correlated with y + p.correct <- plogis(y_bias*scale(y) + qlogis(prediction_accuracy)) + + acc.x0 <- p.correct[df[,x==0]] + acc.x1 <- p.correct[df[,x==1]] + + df[x==0,w:=rlogis(.N,qlogis(1-acc.x0))] + df[x==1,w:=rlogis(.N,qlogis(acc.x1))] + + 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)) + 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=1400, help="number of observations of w") +parser <- add_argument(parser, "--N", default=1000, 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=50, help='seed for the rng') +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=0.01) parser <- add_argument(parser, "--prediction_accuracy", help='how accurate is the predictive model?', default=0.73) 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, "--proxy_formula", help='formula for the proxy variable', default="w_pred~x*y") +parser <- add_argument(parser, "--y_bias", help='coefficient of y on the probability a classification is correct', default=-0.75) args <- parse_args(parser) B0 <- 0 -Bxy <- 0.3 +Bxy <- args$Bxy Bzy <- args$Bzy +Bzx <- args$Bzx if(args$m < args$N){ - df <- simulate_data(args$N, args$m, B0, Bxy, args$Bzx, Bzy, args$seed, args$y_explained_variance, args$prediction_accuracy, args$accuracy_imbalance_difference) - result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, Bzx=args$Bzx, 'Bzy'=Bzy, 'seed'=args$seed, 'y_explained_variance'=args$y_explained_variance, 'prediction_accuracy'=args$prediction_accuracy, 'accuracy_imbalance_difference'=args$accuracy_imbalance_difference, error='') + df <- simulate_data(args$N, args$m, B0, Bxy, Bzx, Bzy, args$seed, args$y_explained_variance, args$prediction_accuracy, y_bias=args$y_bias) + + ## df.pc <- df[,.(x,y,z,w_pred)] + ## # df.pc <- df.pc[,err:=x-w_pred] + ## pc.df <- pc(suffStat=list(C=cor(df.pc),n=nrow(df.pc)),indepTest=gaussCItest,labels=names(df.pc),alpha=0.05) + ## plot(pc.df) + + result <- list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy, Bzx=args$Bzx, 'Bzy'=Bzy, '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$y_bias,error='') - outline <- run_simulation(df, result, outcome_formula=y~x+z, proxy_formula=w_pred~x+z+y+x:y, truth_formula=x~z) + outline <- run_simulation(df, result, outcome_formula=y~x+z, proxy_formula=as.formula(args$proxy_formula), truth_formula=x~z) outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE) if(file.exists(args$outfile)){