Kernel-Distance-Based Covariate Balancing

TL;DR

This paper introduces a kernel-distance-based covariate balancing method to improve causal effect estimation in observational studies by reducing confounding bias through optimal reweighting of treatment groups.

Contribution

The paper proposes a novel preprocessing technique that uses kernel distances to achieve covariate balance, enhancing causal inference in observational data.

Findings

Effective reduction of confounding bias demonstrated in simulations

Improved accuracy in causal effect estimation shown in real data analysis

Method outperforms existing covariate balancing techniques

Abstract

A common concern in observational studies focuses on properly evaluating the causal effect, which usually refers to the average treatment effect or the average treatment effect on the treated. In this paper, we propose a data preprocessing method, the Kernel-distance-based covariate balancing, for observational studies with binary treatments. This proposed method yields a set of unit weights for the treatment and control groups, respectively, such that the reweighted covariate distributions can satisfy a set of pre-specified balance conditions. This preprocessing methodology can effectively reduce confounding bias of subsequent estimation of causal effects. We demonstrate the implementation and performance of Kernel-distance-based covariate balancing with Monte Carlo simulation experiments and a real data analysis.