A0728
Title: Inference on the distribution of individual treatment effects in nonseparable triangular models
Authors: Jun Ma - Renmin University of China (China) [presenting]
Abstract: The focus is on the nonseparable triangular model with binary endogenous treatment and instrumental variables and the rank invariance assumption. It is shown that the empirical distribution function of the estimated individual treatment effects (ITEs) of a prior study converges in distribution to a tight Gaussian process. The asymptotic validity of the nonparametric bootstrap inference is also shown, including both pointwise percentile bootstrap confidence intervals and uniform confidence bands using nonparametric bootstrap critical values.
