Seminario Ingeniera Industrial y Sistemas - P. Universidad Catolica de Chile
Date: 11/20/2024
Abstract
Difference-in-Differences (DD) is a commonly-used approach in policy evaluation where, under a parallel trend assumption, we can recover a causal effect by comparing the difference in outcomes between a treatment and a control group, both before and after an intervention was set in place. However, confounders that differentially vary over time often break the identifying assumption, biasing our estimates and rendering our design invalid. In this paper, I identify contexts where matching can help to eliminate or reduce bias, increasing the robustness of estimates under different sensitivity analyses, and show how balancing covariates directly can yield better results than other forms of adjustment or no adjustment at all. I illustrate these results with simulations and a case study of the impact of a new voucher scheme on socioeconomic segregation in Chile.