Counterfactual Sensitivity and Robustness

We propose a framework for analyzing the sensitivity of counterfactuals to parametric assumptions about distribution latent variables in structural models. In particular, we derive bounds on as spans nonparametric neighborhoods given specification while other “structural” features model are maintained. Our approach recasts infinite‐dimensional problem optimizing counterfactual with respect (subject constraints) finite‐dimensional convex program. also develop an MPEC version our method further simplify computation models endogenous parameters (e.g., value functions) defined by equilibrium constraints. plug‐in estimators and two methods inference. show that converge sharp neighborhood size becomes large. To illustrate broad applicability procedure, present empirical applications matching transferable utility dynamic discrete choice