ConfoundingSHAP: Quantifying confounding strength in causal inference

TL;DR

ConfoundingSHAP is a novel Shapley-based method designed to quantify the confounding strength of individual covariates in observational causal inference, aiding in identifying variables that influence both treatment and outcomes.

Contribution

The paper introduces a new Shapley game tailored for confounding strength attribution and a scalable estimation method that avoids exhaustive refitting, enhancing practical causal analysis.

Findings

ConfoundingSHAP effectively identifies confounders across various datasets.

The method provides more accurate confounder attribution than standard SHAP explanations.

Scalable estimation improves efficiency in confounding strength evaluation.

Abstract

In causal inference, confounders are variables that influence both treatment decisions and outcomes. However, unlike as in randomized clinical trials, the treatment assignment mechanism in observational studies is not known, and it is thus unclear which covariates act as confounders. Here, we aim to generate insight for causal inference and answer: which of the observed covariates act as confounders? We introduce ConfoundingSHAP, a Shapley-based method for attributing confounding strength to individual covariates. Our contributions are twofold. First, we propose a Shapley game targeted to infer the confounding strength of the covariates. Our resulting Shapley values differ from the standard applications of SHAP explanations on causal targets, such as understanding treatment effect heterogeneity, which are ill-suited for our task. Second, as our task requires evaluating the value function over many adjustment sets, we provide a scalable TabPFN-based estimation that avoids exhaustive refitting. We demonstrate the practical value across various datasets, where ConfoundingSHAP provides informative explanations of which observed covariates drive confounding and thereby helps to provide more insight for causal inference in practice.