ConfoundingSHAP: Quantifying confounding strength in causal inference
For researchers and practitioners in causal inference, this provides a novel tool to interpret confounding in observational studies, though the method is incremental as it adapts existing Shapley value concepts to a specific causal inference task.
The paper introduces ConfoundingSHAP, a Shapley-based method to quantify the confounding strength of individual covariates in observational studies, enabling identification of which observed covariates act as confounders. The method uses a scalable TabPFN-based estimation to avoid exhaustive refitting and demonstrates practical value across various datasets.
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.