Estimating Causal Effects with Panel Data

Abstract
We study estimation of causal effects in a panel data setting. We propose a new estimator that combines a flexible model for the potential outcomes based on a low-rank factor structure with unit and time weights intended to upweight units and time periods similar to the treated units and time periods. We find in simulations closely linked to real data sets that the proposed estimator outperforms two-way-fixed-effect/difference-in-differences, synthetic control, matrix completion and synthetic-difference-in-differences estimators.

Guido Imbens is the Applied Econometrics Professor and Professor of Economics at Stanford Graduate School of Business, Stanford University. He is a Member of the Advisory Board of Tinbergen Institute.

Secure your seat by filling in our registration form. Update November 20, 2024: Please be advised that this seminar is currently fully booked. We will add you to the waiting list.