Policy evaluation and learning with partially identified utility under truncation by death
Abstract
Policy evaluation and learning aim to assess and optimize treatment assignment rules based on individual characteristics. A fundamental challenge arises when outcomes are truncated by death, rendering conventional policy utilities undefined. To address this challenge, we study policy evaluation and learning under truncation by death within the principal stratification framework. We propose the survivor average utility and subgroup survival rates for evaluating treatment policies. Under treatment ignorability and monotonicity, the subgroup survival rates are point identified, whereas the survivor average utility is only partially identified through sharp bounds. We then develop semiparametrically efficient estimators for the subgroup survival rates and hybrid estimators for the survivor average utility bounds. Building on this evaluation framework, we formulate a constrained minimax optimization problem for policy learning that minimizes worst-case regret relative to a benchmark policy while requiring the learned policy to achieve a survival rate no lower than that of the benchmark. We show that the optimal policy admits a threshold representation based on two priority scores. We develop an estimation procedure for the optimal policy and establish regret and feasibility guarantees. Simulation studies and an application to the MIMIC-III clinical dataset demonstrate the practical performance of the proposed methods.
Related
- Sequential Certification of Partially Identified Product Targets: Information Limits and Adaptive Sampling
- Symmetric Hermite quadrature-based balanced truncation for learning linear dynamical systems from derivative data
- Population-Level Decision Curve Analysis May Mislead the Evaluation of Prediction Model Usefulness under Subgroup Utility Heterogeneity
- Defense Filters for Robust Learning of Partially Observed Dynamical Systems
- PAC-Bayesian Bounds for Learning Partially Observed Stochastic Linear Time-Invariant State-Space Systems with Inputs and Sub-Gaussian Noise
BibTeX
@misc{huang2026policy,
title = {{Policy evaluation and learning with partially identified utility under truncation by death}},
author = {Zi Huang and Zhichao Jiang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.38783},
eprint = {2609.38783},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.38783},
}
RIS
TY - UNPB
TI - Policy evaluation and learning with partially identified utility under truncation by death
AU - Zi Huang
AU - Zhichao Jiang
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2609.38783
UR - https://arxiv.org/abs/2609.38783
AB - Policy evaluation and learning aim to assess and optimize treatment assignment rules based on individual characteristics. A fundamental challenge arises when outcomes are truncated by death, rendering conventional policy utilities undefined. To address this challenge, we study policy evaluation and learning under truncation by death within the principal stratification framework. We propose the survivor average utility and subgroup survival rates for evaluating treatment policies. Under treatment ignorability and monotonicity, the subgroup survival rates are point identified, whereas the survivor average utility is only partially identified through sharp bounds. We then develop semiparametrically efficient estimators for the subgroup survival rates and hybrid estimators for the survivor average utility bounds. Building on this evaluation framework, we formulate a constrained minimax optimization problem for policy learning that minimizes worst-case regret relative to a benchmark policy while requiring the learned policy to achieve a survival rate no lower than that of the benchmark. We show that the optimal policy admits a threshold representation based on two priority scores. We develop an estimation procedure for the optimal policy and establish regret and feasibility guarantees. Simulation studies and an application to the MIMIC-III clinical dataset demonstrate the practical performance of the proposed methods.
ER -