报告时间：2022年11月16日上午10:00-11:00
报告地点：腾讯会议（会议ID：925-619-737）
报告嘉宾：苗旺
报告题目：Proximal causal inference with application to test-negative design studies of vaccine effectiveness
报告摘要：
Proximal causal inference with application to test-negative design studies of vaccine effectiveness
Miao, Geng, and Tchetgen Tchetgen (2018) proposed a proximal inference approach for adjustment of unmeasured confounding, which broadens the set of structures under which the causal effect can be identified and estimated. In this talk, I will discuss the recent development and its application in the test-negative design (TND) studies. TND has become a standard approach to evaluate vaccine effectiveness in real-world settings, such as Influenza, Rotavirus, Dengue fever, and more recently COVID-19. In a TND study, individuals who experience symptoms and seek care are recruited and tested for the infectious disease which defines cases and controls. TND is subject to various potential biases, including confounding bias and selection bias due to unobserved healthcare seeking behavior. We present a novel approach to identify and estimate vaccine effectiveness in the target population by carefully leveraging proximal inference to account for potential hidden bias in TND studies. We illustrate with an application to study COVID-19 vaccine effectiveness.
个人简介：
苗旺现为北京大学概率统计系研究员， 2008-2017年在北京大学数学科学学院读本科和博士，2017-2018年在哈佛大学生物统计系做博士后研究，2018年入职北京大学光华管理学院，2020年调入数学科学学院。苗旺的研究兴趣包括因果推断，缺失数据分析，及其在生物统计，流行病学，经济学和人工智能研究中的应用，与合作者提出混杂分析的代理推断理论，发展非随机缺失数据的识别性和双稳健估计理论，以及数据融合的半参数理论。个人网页https://www.math.pku.edu.cn/teachers/mwfy/
扫描下方二维码报名↘
所有消息会在两个群中同步通知
请大家不要重复加群~
