学术报告:Generalized martingale difference divergence: Detecting independence in regression and quantile regression with applications in variable screening


报告人:於州教授     华东师范大学

时间:2020年11月1日上午10:30-12:30

地点:数学与信息学院201报告厅


摘要:Martingale difference divergence measures the departure of independence of random vectors in regression and quantile regression. In this paper, we develop a generalized martingale difference divergence (and its correlation) based on symmetric L´evy measures to detect such an independence. We apply the generalized martingale difference correlation as a marginal utility to do high-dimensional variable screening in regression and quantile regression. Both simulation results and real data illustrations show the promising performance of the developed indexes.


报告人简介:

        於州,华东师范大学统计系博士、香港浸会大学数学系、美国威斯康星大学麦迪逊分校统计系博士后,现任华东师范大学经济学部教授、博士生导师,经济学部副主任,教育部“长江学者奖励计划”青年学者,入选上海市青年科技启明星计划、上海高校特聘教授(东方学者)。主要研究方向为高维数据降维。在统计学顶级期刊Annals of Statistics,JASA,Biometrika等发表论文近30篇。获第十届全国统计科研成果奖二等奖。


欢迎广大师生莅临参加!