信息来源：RUC应用统计科学研究中心 发布日期：2026年6月3日

Efficient Estimation of Average Treatment Effect on the Treated under Endogenous Treatment Assignment

主讲人：Jiwei Zhao

Jiwei Zhao is currently an Associate Professor at the University of Wisconsin-Madison. His research interests include semiparametric statistics, the trade-off between efficiency and robustness, domain adaptation and transfer learning, missing data analysis, and causal inference. He also works on developing trustworthy statistical inference methods for AI-predicted data and, more broadly, for synthetic data. His application areas include patient-reported outcomes, clinical trials, real-world evidence and real-world data, survey data, aging, mental health, and cancer. His work has been published in top-tier statistical journals as well as in leading machine learning conferences. His research has been consistently supported by the US National Science Foundation and the National Institutes of Health. Jiwei is now Associate/Action Editor for journals in both statistics and machine learning. He also serves as Area Chair for NeurIPS, ICML and AISTATS. He is an Elected Fellow of the Americal Statistical Association and an Elected Member of the International Statistical Institute.

1

报告信息

时间

2026年6月9日(周二)

10:00

地点

中国人民大学中关村校区明德主楼1016

2

报告摘要

In this paper, we consider estimation of average treatment effect on the treated (ATT), an interpretable and relevant causal estimand to policy makers when treatment assignment is endogenous. By considering shadow variables that are unrelated to the treatment assignment but related to the outcomes of interest, we establish identification of the ATT. Then we focus on efficient estimation of the ATT by characterizing the geometric structure of the likelihood, deriving the semiparametric efficiency bound for ATT estimation and proposing an estimator that can achieve this bound. We rigorously establish the theoretical results of the proposed estimator. The finite sample performance of the proposed estimator is studied through comprehensive simulation studies as well as an application to our motivating study.

“统计大讲堂”由中国人民大学统计学院与应用统计科学研究中心联合主办，旨在搭建学界与业界的交流平台，促进统计理论与实践的深度融合，推动统计学科研创新，服务国家经济社会发展。作为统计学院人才培养体系的核心组成部分，“统计大讲堂”通过系列学术讲座、专题研讨等形式，为统计学子提供前沿知识传授与学术思维训练，营造开放包容的学术氛围，以培养具有国际视野、创新能力和实践精神的复合型统计人才