A Practical Guide to Counterfactual Estimators for Causal Inference with Time-Series Cross-Sectional Data

51 Pages Posted: 10 Apr 2020 Last revised: 22 Jun 2020

See all articles by Licheng Liu

Licheng Liu

Massachusetts Institute of Technology (MIT) - Department of Political Science

Ye Wang

New York University

Yiqing Xu

Stanford University

Date Written: June 21, 2020

Abstract

This paper introduces a simple framework of counterfactual estimators that estimates the average treatment effect on the treated by directly imputing counterfactuals of treated observations in time-series cross-sectional settings with dichotomous treatments. Examples include estimators that recently emerge in the literature, such as the fixed-effect counterfactual estimator, the interactive fixed-effect counterfactual estimator, and the matrix completion estimator. These estimators provide more reliable causal estimates than conventional two-way fixed effects models when the treatment effects are heterogeneous or unobserved time-varying confounders exist. Moreover, we propose two diagnostic tests, an equivalence test and a placebo test, accompanied by visualization tools, to assist researchers to gauge the validity of the no-time-varying-confounder assumption. We illustrate these methods with two empirical examples from political economy and develop an open-source package, \texttt{fect}, in R and Stata to facilitate implementation.

Keywords: counterfactual methods, two-way fixed effects, parallel trends, interactive fixed effects, matrix completion, equivalence test, placebo test, TSCS data, panel data

Suggested Citation

Liu, Licheng and Wang, Ye and Xu, Yiqing, A Practical Guide to Counterfactual Estimators for Causal Inference with Time-Series Cross-Sectional Data (June 21, 2020). Available at SSRN: https://ssrn.com/abstract=3555463 or http://dx.doi.org/10.2139/ssrn.3555463

Licheng Liu

Massachusetts Institute of Technology (MIT) - Department of Political Science ( email )

77 Massachusetts Avenue
Cambridge, MA 02139
United States

Ye Wang

New York University ( email )

Bobst Library, E-resource Acquisitions
20 Cooper Square 3rd Floor
New York, NY 10003-711
United States
6083384999 (Phone)
19104 (Fax)

Yiqing Xu (Contact Author)

Stanford University ( email )

Stanford, CA 94305
United States

HOME PAGE: http://yiqingxu.org

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
190
Abstract Views
641
rank
174,566
PlumX Metrics