A Distributionally-Robust Framework for Nuisance in Causal Effect Estimation
Preprint
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.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Best-effort adaptationAnnals of Mathematics and Artificial Intelligence, 2024
- Causal machine learning for healthcare and precision medicineRoyal Society Open Science, 2022
- Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of PessimismIEEE Transactions on Information Theory, 2022
- Improving imbalanced classification using near-miss instancesExpert Systems with Applications, 2022
- Learning Decomposed Representations for Treatment Effect EstimationIEEE Transactions on Knowledge and Data Engineering, 2022
- Quasi-oracle estimation of heterogeneous treatment effectsBiometrika, 2020
- Causal Inference for Statistics, Social, and Biomedical Sciences2015
- An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational StudiesMultivariate Behavioral Research, 2011
- The Costs of Low Birth WeightThe Quarterly Journal of Economics, 2005
- Constrained Optimization and Lagrange Multiplier Methods1982