Brief

This paper introduces two novel Bayesian methods for semiparametric inference in difference-in-differences (DiD) research designs. The proposed techniques, a semiparametric Bayesian outcome regression and a doubly robust Bayesian procedure, aim to accurately estimate the average treatment effect on the treated (ATT). The authors provide theoretical guarantees, including semiparametric Bernstein-von Mises theorems, and demonstrate the methods' effectiveness through simulations and an empirical application. AI