CHINA·77779193永利(集团)有限公司-Official website

题目：Multiply robust estimation of average treatment effect with missing data

报告人：秦国友 教授（复旦大学）

时间：2021年6月9日 10:00-11:00

地点： 线上腾讯会议（会议号：733 557 452）

Abstract :  Estimation of average treatment effect (ATE) is an important research content in causal inference. When using the observational data to estimate ATE, there are two main challenges including unbalanced covariates and missing outcomes. In this paper, we extend the “double robustness” to “multiple robustness”, which allows multiple models in estimation. The estimator remains consistent when any pair of models for propensity score and selection probability is correctly specified or any model for outcome regression is correctly specified. Under regularity conditions, the asymptotic normality of the estimator is established. The multiply robust estimator has the same large-sample variance as the doubly robust estimator when the correct models for propensity score, selection probability and outcome regression are included in the estimator. Numerical simulations demonstrate the desired properties of the proposed method. Based on the Aerobics Center Longitudinal Study, using the extended robust approach we found a significant association between baseline physical activity and fitness during the three years of follow-up.

报告人简介：秦国友，教授，博士生导师。复旦大学公共卫生学院生物统计学教研室主任。主要从事生物统计学方法学和应用研究，包括针对复杂数据、复杂统计模型的统计方法研究，以及生物统计学方法在医学和公共卫生领域的应用。在医学顶级期刊British Medical Journal（BMJ）和生物统计权威期刊Biometric, Biostatistics, Statistics in Medicine等学术期刊上发表80余篇研究论文。在纵向数据方面相关研究工作获得教育部高等学校科学研究优秀成果奖二等奖。担任中华预防医学会生物统计分会第一届青年委员会主任委员以及<<中国卫生统计>>编委。

联系人：胡涛

举办单位：77779193永利官网、交叉科学研究院、北京应用统计学会

欢迎全体师生积极参加！