Lee’s treatment effect bounds for non-random sample selection - an implementation in Stata

Abstract

Non-random sample selection may render estimated treatment effects biased even if assignment of treatment is purely random. Lee (2009) proposes an estimator for treatment effect bounds that limit the possible range of the treatment effect. In this approach, the lower and upper bound, respectively, correspond to extreme assumptions about the missing information, which are consistent with the observed data. As opposed to conventional parametric approaches to correcting for sample selection bias, Lee's bounds estimator rests on very few assumptions. We introduce the new Stata command leebounds that implements the estimator in Stata. The command allows for several options, such as tightening bounds by the use of covariates.