English  |   学校主页  |   收藏本站



【主 题】 High order conditional distance covariance  with conditional mutual independence

【报告人】 周 望 , 教授


【时 间】 2019年4月13日(星期六)10:00-11:00

【地 点】 上海财经大学统计与管理学院大楼1208会议室

摘 要】We construct a high order conditional distance covariance, which generalizes the notation of conditional distance covariance. The joint conditional distance covariance is defined as a linear combination of conditional distance covariances, which can capture the joint relation of many random vectors given one vector. Furthermore, we develop a new method of conditional independent test based on the joint conditional distance covariance. Simulation results indicate that the proposed method is very effective. We also apply our method to analyze the relationships of $PM_{2.5}$ in five Chinese cities: Beijing, Tianjin, Jinan, Tangshan and Qinhuangdao by Gaussian graphical model.

嘉宾简介】ZHOU Wang, 2004年7月起在新加坡国立大学统计系任教,并于2009年1月获终身教授。现为新加坡国立大学正教授。 主要研究方向为: random matrices, SLE, high dimensional statistics。近年来发表有较高学术水平的论文五十多篇。 其中在概率统计学方面的国际公认的顶尖杂志Annals of Statistics, Journal of American Statistical Association, Biometrika, Annals of  Probability, Probability Theory and Related Fields, Annals of Applied Probability上发表论文十余篇。2012获得国际统计学会当选成员(Elected Member of International Statistical Institute)