Machine Learning for Difference-in-Differences with Staggered Treatment Timing and Dynamic Treatment Effect Heterogeneity

Authors: Julia Hatamyar, Noemi Kreif, Martin Huber, Rudi Rocha

Abstract:
Using machine learning, we combine the strengths of two recently proposed nonparametric difference-in-differences methods, extending them to enable the examination of treatment effect heterogeneity in the staggered adoption setting. The method allows for estimation of time-varying conditional average treatment effects, which can be used to conduct detailed inference on drivers of treatment effect heterogeneity. We perform simulations to evaluate the performance of the proposed method and use it to evaluate the heterogeneous impacts of Brazil's Family Health Program on infant mortality.