Discussant: Hao Wang, Graduate Student, Department of Statistics, Iowa State University

Title: High-dimensional Covariate Balancing Propensity Score (CBPS)

Abstract: The paper under discussion proposes a new method for estimating Covariate Balancing Propensity Score (CBPS) in high-dimensional settings, where the number of confounders is much larger than sample size. The appealing properties of CBPS remain and the proposed Horvitz–Thompson estimator for E[Yi(1)] is root-n consistent and asymptotically normal so long as either the propensity score model or the outcome model is correctly specified. Besides the nice theoretical properties, I will also discuss empirical performance through some numeric studies.