Efficient nonparametric estimation of causal mediation effects

TL;DR

This paper introduces a new nonparametric method for causal mediation analysis that avoids parametric assumptions, providing efficient estimation of direct and indirect effects and extending to multiple mediators.

Contribution

It develops a globally semiparametric efficient estimator for causal mediation effects that does not require modeling multiple conditional distributions.

Findings

Estimator is globally semiparametric efficient

Consistent variance estimation method provided

Method extended to multiple mediators

Abstract

An essential goal of program evaluation and scientific research is the investigation of causal mechanisms. Over the past several decades, causal mediation analysis has been used in medical and social sciences to decompose the treatment effect into the natural direct and indirect effects. However, all of the existing mediation analysis methods rely on parametric modeling assumptions in one way or another, typically requiring researchers to specify multiple regression models involving the treatment, mediator, outcome, and pre-treatment confounders. To overcome this limitation, we propose a novel nonparametric estimation method for causal mediation analysis that eliminates the need for applied researchers to model multiple conditional distributions. The proposed method balances a certain set of empirical moments between the treatment and control groups by weighting each observation; in particular, we establish that the proposed estimator is globally semiparametric efficient. We also show how to consistently estimate the asymptotic variance of the proposed estimator without additional efforts. Finally, we extend the proposed method to other relevant settings including the causal mediation analysis with multiple mediators.