Home  >>  Archives  >>  Volume 14 Number 4  >>  st0364

The Stata Journal
Volume 14 Number 4: pp. 884-894



Subscribe to the Stata Journal
cover

Lee (2009) treatment-effect bounds for nonrandom sample selection

Harald Tauchmann
University of Erlangen-Nuremberg,
Rheinisch-Westfälisches Institut für Wirtschaftsforschung (RWI),
and Centre of Health Economics Research (CINCH)
Nürnberg, Germany
harald.tauchmann@fau.de
Abstract.  Nonrandom sample selection may render estimated treatment effects biased even if assignment of treatment is purely random. Lee (2009, Review of Economic Studies, 76: 1071–1102) proposes an estimator for treatment-effect bounds that limit the possible range of the treatment effect. In this approach, the lower and upper bound correspond to extreme assumptions about the missing information that are consistent with the observed data. In contrast to conventional parametric approaches to correcting for sample-selection bias, Lee’s bounds estimator rests on very few assumptions. I introduce the new command leebounds, which implements the estimator in Stata. The command allows for several options, such as tightening bounds by using covariates.
Terms of use     View this article (PDF)

View all articles by this author: Harald Tauchmann

View all articles with these keywords: leebounds, nonparametric, randomized trial, sample selection, attrition, bounds, treatment effect

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS