Forecasting Algorithms for Causal Inference with Panel Data

arXiv:2208.03489v3 Announce Type: replace-cross
Abstract: Conducting causal inference with panel data is a core challenge in social science research. We adapt a deep neural architecture for time series forecasting (the N-BEATS algorithm) to more accurately impute the counterfactual evolution of a treated unit had treatment not occurred. Across a range of settings, the resulting estimator (``SyNBEATS'') significantly outperforms commonly employed methods (synthetic controls, two-way fixed effects), and attains comparable or more accurate performance compared to recently proposed methods (synthetic difference-in-differences, …

abstract adapt algorithm algorithms architecture arxiv causal causal inference challenge core counterfactual data econ.em evolution forecasting inference panel research science series social social science stat.ml time series time series forecasting treatment type