大发棋牌236ccios|首頁|歡迎您

Adjusting the Benjamini-Hochberg method for controlling the false discovery rate in knockoff-assisted variable selection

时间:2022-05-17         阅读:

光华讲坛——社会名流与企业家论坛第6134


主题Adjusting the Benjamini-Hochberg method for controlling the false discovery rate in knockoff-assisted variable selection

主讲人美国天普大学 汤琤咏教授

主持人统计学院 常晋源教授

时间2022年5月20日(周五)上午9:00-10:00

举办地点:腾讯会议,863 519 982

主办单位:数据科学与商业智能联合实验室 统计学院 科研处

主讲人简介:

Dr. Cheng Yong Tang is Professor and the Cyrus C.K. Curtis Senior Research Fellow in the Department of Statistics, Operations, and Data Science in the Fox School of Business of Temple University.

He is an Associate Editor of Journal of the American Statistical Association, Theory and Methods, an Associate Editor for Reproducibility of Journal of the American Statistical Association, Applications and Case Studies, an Associate Editor of Journal of Business and Economic Statistics, an Associate Editor of Statistica Sinica, and an Associate Editor of Computational Statistics and Data Analysis. He served as the Director of the Graduate Programs in Statistics of the Department of Statistical Science in 2016-2019. Dr Tang is an Elected Member of the International Statistical Institute, a member of the American Statistical Association, a member of the Institute of Mathematical Statistics, and a member of the International Chinese Statistical Association.

Dr. Tang earned his Ph.D. in Statistics from the Department of Statistics, Iowa State University. This is his Mathematical Genealogy. His research interests are methods, theory, and applications in statistics and data science.

汤琤咏,天普大学福克斯商学院统计、运营和数据科学系教授、Cyrus C.K. Curtis高级研究员。

他是Journal of the American Statistical Association理论与方法领域和应用、案例研究领域的副主编,Journal of Business and Economic Statistics的副主编,Statistica Sinica的副主编,以及Computational Statistics and Data Analysis的副主编。2016-2019年,他担任统计科学系统计学研究生项目主任。汤教授是国际统计学会的当选会员,美国统计学会的会员,国际数理统计学会的会员和泛华统计协会会员。

汤教授博士毕业于爱荷华州立大学统计系。他的研究兴趣是统计学和数据科学的方法、理论和应用。

内容简介

The knockoff-based multiple testing setup of Barber and Candes (2015) for variable selection in multiple regression where sample size is as large as the number of explanatory variables is considered. The method of Benjamini and Hochberg (1995) based on ordinary least squares estimates of the regression coefficients is adjusted to this setup, transforming it to a valid p-value based false discovery rate controlling method not relying on any specific correlation structure of the explanatory variables. Simulations and real data applications show that our proposed method that is agnostic to π0 , the proportion of unim-portant explanatory variables, and a data-adaptive version of it that uses an estimate of π0 are powerful competitors of the false discovery rate controlling method in Barber and Candes (2015). This is a joint work with Dr Sanat K. Sarkar.

考虑到Barber和Candes(2015)在样本量与解释变量数量相同的多重回归中变量选择的仿品多重测试设置,本文将Benjamini和Hochberg(1995)基于回归系数普通最小二乘估计的方法调整为这种设置,将其转化为不依赖于解释变量的任何特定相关结构的基于p值的有效错误发现率控制方法。模拟和实际数据应用表明,本文提出的方法不依赖于π0,单一重要解释变量的比例,以及使用π0估计的数据自适应版本,是Barber和Candes(2015)提出的错误发现率控制方法的有力竞争对手。这是与Sanat K. Sarkar博士的联合研究。