Benchmarking Bayesian Causal Discovery Methods for Downstream Treatment Effect Estimation

TL;DR

This paper evaluates how well various causal discovery methods perform in estimating treatment effects, highlighting their strengths and weaknesses across synthetic and real-world data, especially in low-data scenarios.

Contribution

It introduces a distribution-level evaluation framework for causal discovery methods applied to treatment effect estimation, including a new GFlowNet-based method.

Findings

Some algorithms effectively capture diverse ATE modes

Certain methods learn many low-probability modes affecting recall and precision

Evaluation across synthetic and real-world scenarios reveals varied performance

Abstract

The practical utility of causality in decision-making is widespread and brought about by the intertwining of causal discovery and causal inference. Nevertheless, a notable gap exists in the evaluation of causal discovery methods, where insufficient emphasis is placed on downstream inference. To address this gap, we evaluate seven established baseline causal discovery methods including a newly proposed method based on GFlowNets, on the downstream task of treatment effect estimation. Through the implementation of a distribution-level evaluation, we offer valuable and unique insights into the efficacy of these causal discovery methods for treatment effect estimation, considering both synthetic and real-world scenarios, as well as low-data scenarios. The results of our study demonstrate that some of the algorithms studied are able to effectively capture a wide range of useful and diverse ATE modes, while some tend to learn many low-probability modes which impacts the (unrelaxed) recall and precision.