LongBet: Heterogeneous Treatment Effect Estimation in Panel Data

TL;DR

This paper presents LongBet, a Bayesian semi-parametric method for estimating heterogeneous treatment effects in panel data without relying on the parallel trend assumption, effectively handling short panels with observed confoundings.

Contribution

It introduces a novel Bayesian causal forest model tailored for panel data that estimates treatment effects without the parallel trend assumption and quantifies uncertainty.

Findings

Performs well in simulations with and without parallel trends

Enables estimation of conditional average treatment effects

Provides uncertainty quantification for estimates

Abstract

This paper introduces a novel approach for estimating heterogeneous treatment effects of binary treatment in panel data, particularly focusing on short panel data with large cross-sectional data and observed confoundings. In contrast to traditional literature in difference-in-differences method that often relies on the parallel trend assumption, our proposed model does not necessitate such an assumption. Instead, it leverages observed confoundings to impute potential outcomes and identify treatment effects. The method presented is a Bayesian semi-parametric approach based on the Bayesian causal forest model, which is extended here to suit panel data settings. The approach offers the advantage of the Bayesian approach to provides uncertainty quantification on the estimates. Simulation studies demonstrate its performance with and without the presence of parallel trend. Additionally, our proposed model enables the estimation of conditional average treatment effects, a capability that is rarely available in panel data settings.