Exact Trend Control in Estimating Treatment Effects Using Panel Data with Heterogenous Trends

TL;DR

This paper introduces exact-matching methods to estimate treatment effects in panel data models with heterogeneous trends, effectively removing confounding effects and providing more accurate estimates, as demonstrated in a case study on California's tobacco control program.

Contribution

The paper proposes novel exact-matching techniques that eliminate heterogenous trend confounding in panel data treatment effect estimation.

Findings

New estimators suggest smaller effects of tobacco control program

Methods effectively control for heterogenous trends

Application to real data demonstrates improved accuracy

Abstract

For a panel model considered by Abadie et al. (2010), the counterfactual outcomes constructed by Abadie et al., Hsiao et al. (2012), and Doudchenko and Imbens (2017) may all be confounded by uncontrolled heterogenous trends. Based on exact-matching on the trend predictors, I propose new methods of estimating the model-specific treatment effects, which are free from heterogenous trends. When applied to Abadie et al.'s (2010) model and data, the new estimators suggest considerably smaller effects of California's tobacco control program.