This R package implements sharp bounds on treatments effects in the presence of selection/nonresponse bias in randomized control trials. It includes basic Lee bounds Lee (2009) and sharp bounds based on machine learning Semenova(2020). See vignette leebounds for description of the package (available through vignette("leebounds") once package is installed), and the package manual for documentation of the package functions.
install.packages("devtools")
library(devtools)
install_github("vsemenova/leebounds")
library(leebounds)
# compute basic Lee (2009) bounds for ATE in week 208
leedata=data.frame(treat=JobCorps_baseline$TREATMNT.y,selection=JobCorps_employment$week_208,outcome=JobCorps_wages$week_208)
GetBounds(leebounds(leedata))
Vira Semenova: semenovavira@gmail.com