Simulation-based sensitivity analysis for matching estimators

This article presents a Stata program (sensatt) that implements the sensitivity analysis for matching estimators proposed by Ichino, Mealli, and Nannicini (Journal of Applied Econometrics , forthcoming). The analysis simulates a potential confounder to assess the robustness of the estimated treatment effects with respect to deviations from the conditional independence assumption. The program uses the commands for propensity-score matching (att* ) developed by Becker and Ichino (Stata Journal 2: 358–377). I give an example by using the National Supported Work demonstration, widely known in the program evaluation literature. Copyright 2007 by StataCorp LP.