Variable Selection in Causal Inference Using Penalization

TL;DR

This paper introduces a penalized likelihood method for variable selection in causal inference that considers both outcome and treatment models simultaneously, improving confounder selection in high-dimensional data.

Contribution

It proposes a novel variable selection approach that attains the oracle property and enhances causal effect estimation in high-dimensional settings.

Findings

Method attains oracle property under certain conditions.

Improves confounder selection accuracy.

Demonstrated effectiveness on economic growth data.

Abstract

In the causal adjustment setting, variable selection techniques based on either the outcome or treatment allocation model can result in the omission of confounders or the inclusion of spurious variables in the propensity score. We propose a variable selection method based on a penalized likelihood which considers the response and treatment assignment models simultaneously. The proposed method facilitates confounder selection in high-dimensional settings. We show that under some conditions our method attains the oracle property. The selected variables are used to form a double robust regression estimator of the treatment effect. Simulation results are presented and economic growth data are analyzed.