时 间：2023年12月5日 9:30-10:30
地 点：腾讯会议ID：489-888-851
报告人：李伟中国人民大学副教授
主持人：马慧娟 华东师范大学副教授
摘 要：
Causal mediation analysis aims to investigate how an intermediary factor, called a mediator, regulates the causal effect of a treatment on an outcome. With the increasing availability of measurements on a large number of potential mediators, various methods for mediator selection analysis have been proposed. However, these methods often assume the absence of unmeasured mediator-outcome confounding. We allow for such confounding and offer a linear structural equation modeling framework to tackle the mediator selection issue. To achieve this, we first identify causal parameters by constructing a pseudo proxy variable for unmeasured confounding. Leveraging this proxy variable, we propose a partially penalized method to identify mediators affecting the outcome. The resultant estimates are consistent, and the estimates of nonzero parameters are asymptotically normal. Furthermore, we introduce a two-step procedure to consistently select active mediation pathways, eliminating the need to test composite null hypotheses for each mediator, as commonly required by traditional methods. Simulation studies demonstrate the superior performance of our approach compared to existing methods. Finally, we apply our approach to genomic data, identifying gene expressions that potentially mediate the impact of a genetic variant on mouse obesity.
报告人简介：
李伟，中国人民大学统计学院副教授，研究方向是缺失数据、因果推断、高维统计及其在生物医学、社会经济学等领域中的应用，已在JRSSB, Biometrika, Journal of Econometrics, Biometrics等国际统计学期刊发表多篇文章。主持国家自然科学青年基金、国家重点研发计划青年科学家项目子课题、北京市自然科学基金面上项目、全国统计科学研究重点项目等多项课题，担任现场统计学会因果推断分会副秘书长和北京生物医学统计与数据管理研究会理事。