leebounds: Estimating Lee (2009) treatment effect bounds

In vsemenova/leebounds: Nonparametric bounds on Average Treatment Effect in the presence of Endogenous Selection

Estimating Lee (2009) treatment effect bounds

Description

leebounds basic Lee (2009) bounds on treatment effect without covariates. Bounds are defined under monotonicity assumption stating that treatment cannot hurt selection.

Usage

leebounds(leedata)

Arguments

leedata

data frame containing three fields \itemresleedata$treat: binary treatment indicator \itemresleedata$selection: selection=1 if the outcome is observed \itemresleedata$outcome: outcome=selection*outcome

Value

A list containing the estimate of lower bound and upper bound

References

David Lee (2009). Training, Wages, and Sample Selection: Estimating Sharp Bounds on Treatment Effects. The Review of Economic Studies, 76(3) 1071-1102. https://www.princeton.edu/~davidlee/wp/resrevision8.pdf

Examples

n <- 500; x <- matrix(rnorm(n*5),nrow=n)
a <- runif(n); y <- a + rnorm(n,sd=.5)

ce.res <- ctseff(y,a,x, bw.seq=seq(.2,2,length.out=100))
plot.ctseff(ce.res)

# check that bandwidth choice is minimizer
plot(ce.res$bw.risk$bw,ce.res$bw.risk$risk)

vsemenova/leebounds documentation built on Sept. 30, 2023, 8:30 a.m.