X-Git-Url: https://code.communitydata.science/ml_measurement_error_public.git/blobdiff_plain/6057688060b5bf2a94f2b96b65b275a91991c0f3..e41d11afb9a80180feff844666e3ee463d20a7cd:/simulations/example_2.R

diff --git a/simulations/example_2.R b/simulations/example_2.R
new file mode 100644
index 0000000..2ec23f7
--- /dev/null
+++ b/simulations/example_2.R
@@ -0,0 +1,87 @@
+### EXAMPLE 2: demonstrates how measurement error can lead to a type sign error in a covariate
+### 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)
+
+## 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)
+#### 
+source("simulation_base.R")
+
+simulate_latent_cocause <- function(N, m, B0, Bxy, Bgy, Bkx, Bgx, seed){
+    set.seed(seed)
+
+    ## the true value of x
+
+    g <- rbinom(N, 1, 0.5)
+    k <- rnorm(N, 0, 1)
+    xprime <- Bkx*k + Bgx * g + rnorm(N,0,1)
+    xvec <- scale(xprime)
+
+    y <-  Bxy * xvec  + Bgy * g + rnorm(N, 0, 1) + B0
+
+    df <- data.table(x=xvec,k=k,y=y,g=g)
+    names(df) <- c('x','k','y','g')
+    if( m < N){
+        df <- df[sample(nrow(df), m), x.obs := x]
+    } else {
+        df <- df[, x.obs := x]
+    }
+
+    w.model <- lm(x ~ k,df)
+    w <- predict(w.model,data.frame(k=k)) 
+    w <- logistic(w + rnorm(N,0,sd(w)*0.1))
+    ## y = B0 + B1x + e
+
+    df[,':='(w=w, w_pred = as.integer(w>0.5))]
+    return(df)
+}
+
+
+parser <- arg_parser("Simulate data and fit corrected models")
+parser <- add_argument(parser, "--N", default=1000, help="number of observations of w")
+parser <- add_argument(parser, "--m", default=200, help="m the number of ground truth observations")
+parser <- add_argument(parser, "--seed", default=4321, help='seed for the rng')
+parser <- add_argument(parser, "--outfile", help='output file', default='example_2.feather')
+args <- parse_args(parser)
+
+Ns <- c(1000, 10000, 1e6)
+ms <- c(100, 250, 500, 1000)
+B0 <- 0
+Bxy <- 0.2
+Bgy <- -0.2
+Bkx <- 2
+Bgx <- 3
+
+outline <- run_simulation(simulate_latent_cocause(args$N, args$m, B0, Bxy, Bgy, Bkx, Bgx, args$seed)
+                         ,list('N'=args$N,'m'=args$m,'B0'=B0,'Bxy'=Bxy,'Bgy'=Bgy, 'Bkx'=Bkx, 'Bgx'=Bgx, 'seed'=args$seed))
+
+outfile_lock <- lock(paste0(args$outfile, '_lock'),exclusive=TRUE)
+if(file.exists(args$outfile)){
+    logdata <- read_feather(args$outfile)
+    logdata <- rbind(logdata,as.data.table(outline))
+} else {
+    logdata <- as.data.table(outline)
+}
+
+print(outline)
+write_feather(logdata, args$outfile)
+unlock(outfile_lock)