When Should We Use Unit Fixed Effects Regression Models for Causal Inference with Longitudinal Data?

2019/03/07 by Kosuke Imai, In Song Kim · 1 voice · 25 citations
Mathematics · #Advanced Causal Inference Techniques #Statistical Methods and Bayesian Inference #Statistical Methods and Inference

paper · doi:10.1111/ajps.12417

openalex publication_date 2019/03/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30

Abstract

Abstract Many researchers use unit fixed effects regression models as their default methods for causal inference with longitudinal data. We show that the ability of these models to adjust for unobserved time‐invariant confounders comes at the expense of dynamic causal relationships, which are permitted under an alternative selection‐on‐observables approach. Using the nonparametric directed acyclic graph, we highlight two key causal identification assumptions of unit fixed effects models: Past treatments do not directly influence current outcome, and past outcomes do not affect current treatment. Furthermore, we introduce a new nonparametric matching framework that elucidates how various unit fixed effects models implicitly compare treated and control observations to draw causal inference. By establishing the equivalence between matching and weighted unit fixed effects estimators, this framework enables a diverse set of identification strategies to adjust for unobservables in the absence of dynamic causal relationships between treatment and outcome variables. We illustrate the proposed methodology through its application to the estimation of GATT membership effects on dyadic trade volume.

Citations

• On the Validity of Covariate Adjustment for Estimating Causal Effects
• A Theorem on the Manipulability of Redundant Serial Kinematic Chains
• Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations
• A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity
• What Are We Weighting For?

Cited by

• Adjusting for Unmeasured Confounding in Marginal Structural Models with Propensity-Score Fixed Effects
• Fixed Effects and the Generalized Mundlak Estimator
• The promise and perils of the sharing economy: The impact of Airbnb lettings on crime
• Causal effect estimates of online e-cigarette marketing exposure on future e-cigarette harm perception and use
• Causal inference on human behaviour
• A causal framework for the drivers of animal social network structure
• Suppressing Black Votes: A Historical Case Study of Voting Restrictions in Louisiana
• These Are Not the Effects You Are Looking for: Causality and the Within-/Between-Persons Distinction in Longitudinal Data Analysis
• The impact of carbon emission trading schemes on urban-rural income inequality in China: A multi-period difference-in-differences method
• Using Multiple Pre-treatment Periods to Improve Difference-in-Differences and Staggered Adoption Design
• Learning About the Binding Nature of the Law: Police Violence, Criminal Offending and Adolescent Legal Socialization
• Varieties of Capitalism and Cross-national Variation in Fertility Rates
• Working-class structural power, associational power, and income inequality
• Bridging Structural Causal Inference and Machine Learning The S-DIDML Estimator for Heterogeneous Treatment Effects
• Assessing Outcome-to-Outcome Interference in Sibling Fixed Effects Models
• Causal Panel Analysis under Parallel Trends: Lessons from a Large Reanalysis Study
• Aggressive policing and undermined legitimacy: assessing the impact of police stops at gunpoint on perceptions of police in São Paulo, Brazil
• From ideal experiments to ideal research designs (IDRs): What they are and why we should use them more
• Estimating Sibling Spillover Effects with Unobserved Confounding Using Gain-Scores
• On the Use of Two-Way Fixed Effects Regression Models for Causal Inference with Panel Data
• Labor market risks and welfare preferences: a bounded rationality approach
• The Right to Work, Power Resources, and Economic Inequality
• Mobilize for Our Lives? School Shootings and Democratic Accountability in U.S. Elections
• Using Small Domain Estimation to obtain better retrospective Age Period Cohort insights
• Using PanelMatch to Estimate the Causal Effects of Police Interventions with Irregular Treatment Patterns: An Assessment of the NYPD’s Summer All Out, 2014–2019

Discussions

• . @igelstorm.bsky.social, I attended your talk at the 2025 Turing CIIG Seminar and loved it. I wanted to share this article by Imai et al. (2019), which provides estimators when data on exposure is ar [bsky, 1 points, 0 comments]

Related

• When Do Treatment Changes Identify Causal Effects?
• Causal Models for Longitudinal and Panel Data: A Survey
• Causal models for longitudinal and panel data: a survey
• Dynamic Treatment Effects under Functional Longitudinal Studies
• On the Use of Two-Way Fixed Effects Regression Models for Causal Inference with Panel Data