Forecasting Algorithms for Causal Inference with Panel Data

TL;DR

This paper introduces SyNBEATS, a neural network-based estimator for causal inference with panel data, which improves counterfactual predictions over traditional methods by leveraging advanced time series forecasting techniques.

Contribution

It adapts the N-BEATS neural architecture for causal inference, demonstrating superior performance over existing methods in panel data settings.

Findings

SyNBEATS outperforms synthetic controls and fixed effects.

Achieves comparable or better results than recent methods like synthetic difference-in-differences.

Provides an open-source implementation for practical use.

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, matrix completion). An implementation of this estimator is available for public use. Our results highlight how advances in the forecasting literature can be harnessed to improve causal inference in panel data settings.