报告人：徐功军
报告地点：数学与统计学院415室
报告时间：2018年06月29日星期五14:00-16:00
邀请人：
报告摘要：
Many high dimensional hypothesis tests examine the moments of the distributions that are of interest, such as testing of mean vectors and covariance matrices. We propose a general framework that constructs a family of U statistics as unbiased estimators of those moments. The usage of the framework is illustrated by testing off-diagonal elements of a covariance matrix. We show that under null hypothesis, the U statistics of different finite orders are asymptotically independent and normally distributed. Moreover, they are also asymptotically independent with the max-type test statistic. Based on the asymptotic independence property, we further construct an adaptive testing procedure that maintains high power across a wide range of alternatives.
主讲人简介：
Ph.D. in Statistics, Columbia University 05/2013 Assistant Professor of Statistics, University of Minnesota 08/2013 – 12/2016 Assistant Professor of Statistics, University of Michigan 01/2017 – Present Assistant Professor of Psychology (by courtesy), University of Michigan 01/2017 – Present
