A0862
Title: Universal difference-in-differences
Authors: Chan Park - University of Pennsylvania (United States) [presenting]
Eric Tchetgen Tchetgen - The Wharton School, University of Pennsylvania (United States)
Abstract: Difference-in-differences (DiD) is a popular method for evaluating real-world policy interventions' causal effects. DiD relies on the parallel trends (PT) assumption to identify the average treatment effect on the treated, which states that the time trends for the average treatment-free potential outcomes are parallel across the treated and control groups. A well-known limitation of the PT assumption is its lack of generalization to causal effects for discrete outcomes and to nonlinear effect measures. Universal Difference-in-Differences (UDiD) based on an alternative assumption to PT is considered for identifying treatment effects for the treated on any scale of potential interest and outcomes of an arbitrary nature. Specifically, the odds ratio equi-confounding (OREC) assumption is introduced, which states that the generalized odds ratios relating the treatment-free potential outcome and treatment are equivalent across time periods. Under the OREC assumption, nonparametric identification for any potential treatment effect on the treated in view is established. Moreover, a consistent, asymptotically linear, and semiparametric efficient estimator is developed for any given treatment effect on the treatment of interest, which leverages recent learning theory. UDiD with simulations and two real-world applications in labour economics and traffic safety evaluation are illustrated.
