Double machine learning for static panel models with fixed effects

Abstract

Summary Recent advances in causal inference have seen the development of methods that make use of the predictive power of machine learning algorithms. In this paper, we develop novel double machine learning procedures for panel data in which these algorithms are used to approximate high-dimensional and non-linear nuisance functions of the covariates. Our new procedures are extensions of the well-known correlated random effects, within-group, and first-difference estimators from linear to non-linear panel models, specifically, the partially linear regression model with fixed effects and unspecified non-linear confounding. Our simulation study assesses the performance of these procedures using different machine learning algorithms. We use our procedures to re-estimate the impact of the introduction of the National Minimum Wage on voting behaviour in the United Kingdom. From our results, we recommend the use of first-differencing because it imposes the fewest constraints on the distribution of the fixed effects, and an ensemble learning strategy to ensure optimum estimator accuracy.