Treatment Effect Estimation for Optimal Decision-Making

TL;DR

This paper reveals that state-of-the-art two-stage CATE estimators may be suboptimal for decision-making and introduces a new method to improve their decision performance through a retargeted learning objective.

Contribution

It proposes a novel two-stage learning objective that aligns CATE estimation with decision-making goals and introduces a neural method to optimize this objective.

Findings

The new method improves decision-making accuracy in experiments.

Theoretical analysis shows better alignment with decision performance.

Empirical results confirm the effectiveness of the proposed approach.

Abstract

Decision-making across various fields, such as medicine, heavily relies on conditional average treatment effects (CATEs). Practitioners commonly make decisions by checking whether the estimated CATE is positive, even though the decision-making performance of modern CATE estimators is poorly understood from a theoretical perspective. In this paper, we study optimal decision-making based on two-stage CATE estimators (e.g., DR-learner), which are considered state-of-the-art and widely used in practice. We prove that, while such estimators may be optimal for estimating CATE, they can be suboptimal when used for decision-making. Intuitively, this occurs because such estimators prioritize CATE accuracy in regions far away from the decision boundary, which is ultimately irrelevant to decision-making. As a remedy, we propose a novel two-stage learning objective that retargets the CATE to balance CATE estimation error and decision performance. We then propose a neural method that optimizes an adaptively-smoothed approximation of our learning objective. Finally, we confirm the effectiveness of our method both empirically and theoretically. In sum, our work is the first to show how two-stage CATE estimators can be adapted for optimal decision-making.