今日云讲堂 | 苗旺:Introducing the specificity score
报告信息
主题:
Introducing the specificity score: a measure of causality beyond P value
嘉宾:苗旺
地点:腾讯会议:798-676-334 (或点击阅读原文)
时间:2023年06月10日(周六) 19:00
嘉宾简介
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,半参数统计,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,获得国家重点研发计划青年科学家项目和国家自然科学基金面上项目资助。担任中国现场统计研究会因果推断分会常务副理事长。
个人网页:
https://www.math.pku.edu.cn/teachers/mwfy
报告摘要
There is considerable debate and doubt about the use of P value in scientific research in recent years, particularly after its use is banished in several prestigious journals. Much scientific research is concerned with uncovering causal associations, however, P value is mostly a measure of the significance of a statistical association, which could be biased from the causal association of interest and lead to false/trivial scientific discoveries particularly in the presence of unmeasured confounding. In this talk, I will introduce a score measuring the specificity of causal associations and a specificity score-based test about the existence of causal effects in the presence of unmeasured confounding. Under certain conditions, this approach has controlled type I error and power approaching unity for testing the null hypothesis of no causal effect. A visualization approach using a heatmap of specificity is proposed to communicate all specificity score/test information in a universal and effective manner. This approach only entails a rough idea on the broadness of the causal associations in sight, e.g., the maximum or upper-bound number of causes/outcomes of an outcome/treatment, but does not require to know exactly the exclusion of certain causal effects or the availability of auxiliary variables. This approach is related to Hill's specificity criterion for causal inference, but I will discuss the difference from Hill's. This approach admits for joint causal discovery with multiple treatments and multiple outcomes, which is particularly suitable for gene expressions studies, Mendelian randomization and EHR studies. Identification and estimation will be briefly covered. Simulations are used for illustration and an application to a mouse obesity dataset detects potential active effects of genes on clinical traits that are relevant to metabolic syndrome.
统计之都(Capital of Statistics,简称 COS)成立于 2006 年,是一个旨在推广与应用统计学、数据科学知识的公益性网站和社区。
统计之都以专业、人本、正直、团结的理念尝试推动统计和数据科学在中国的发展,促进各行业的创新和繁荣。
微信扫码关注该文公众号作者