A0648
Title: Time-varying heterogeneous treatment effects in event studies
Authors: Laura Liu - Indiana University Bloomington (United States) [presenting]
Irene Botosaru - McMaster University (Canada)
Abstract: The identification and estimation of heterogeneous treatment effects are studied in event studies. The importance of both heterogeneity in treatment effects and the inclusion of lagged dependent variables are highlighted. Omitting lagged dependent variables can lead to omitted variable bias in the estimation of time-varying treatment effects. Under the assumption of strict exogeneity in the treatment, an empirical Bayes estimator is proposed for the heterogeneous treatment effects, which is flexible and easy to implement. The method also helps shed light on common assumptions in the event study literature, such as the potential correlation between heterogeneous treatment effects and individual heterogeneity, as well as the potential presence of state dependence.
