A Distributionally Robust Framework for Nuisance in Causal Effect Estimation

TL;DR

This paper introduces a distributionally robust framework that improves causal effect estimation by addressing propensity ambiguity and instability through adversarial loss and regularization, showing consistent improvements on various datasets.

Contribution

It proposes a novel adversarial and regularized approach to handle distribution shift and propensity estimation issues in causal inference.

Findings

Improved causal effect estimation accuracy on synthetic datasets.

Enhanced robustness to propensity score misspecification.

Consistent performance gains on real-world datasets.

Abstract

Causal inference requires evaluating models on balanced distributions between treatment and control groups, while training data often exhibits imbalance due to historical decision-making policies. Most conventional statistical methods address this distribution shift through inverse probability weighting (IPW), which requires estimating propensity scores as an intermediate step. These methods face two key challenges: inaccurate propensity estimation and instability from extreme weights. We decompose the generalization error to isolate these issues--propensity ambiguity and statistical instability--and address them through an adversarial loss function. Our approach combines distributionally robust optimization for handling propensity uncertainty with weight regularization based on weighted Rademacher complexity. Experiments on synthetic and real-world datasets demonstrate consistent improvements over existing methods.