报告人：朱仲义教授,

复旦大学统计系

报告题目：Robust subgroup identification

报告时间：2017年04月21日09:30

报告地点：海韵实验楼108

报告摘要：In many applications, subgroups with different parameters may exist even after accounting for the covariate effects, and it is important to identify the meaningful subgroups for better medical treatment or market segmentation.We propose a robust subgroup identification method based on median regression with concave fusion penalization. The proposed method can simultaneously determine the number of subgroups, identify the group membership for each subject, and estimate the regression coefficients. Without requiring any parametric distributional assumptions, the proposed method is robust against both outliers in the response and heteroscedantcity in the regression error. We develop a convenient algorithm based on local linear approximation and propose a new divide-and-conquer algorithm to alleviate the computing challenge for large samples. In addition, we establish the oracle property of the proposed penalized estimator by employing a recently developed convex-differencing tool to account for the nonsmooth loss and nonconvex penalty functions. The numerical performance of the proposed method is assessed through simulation and the analysis of a heart disease data..

报告人简介：朱仲义，复旦大学统计系教授，博士研究生导师；曾任中国概率统计学会第八、九届副理事长，国际著名杂志”Statistica Sinica”副主编，现任“应用概率统计”，”数理统计与管理”杂志编委, 中国现场统计研究会常务理事，中国统计教材编审委员会委员。专业研究方向为：保险精算；纵向数据(面板数据)模型；分位数回归模型等。主持完成国家自然科学基金四项、国家社会科学基金一项，作为子项目负责人完成国家自然科学基金重点项目一项。目前主持国家自然科学基金重大项目子项目一项，国家自然科学基金一项。近几年发表论文90多篇(其中包括在国际顶级刊物：J.R.Stat.Soc B, J.A.S.A., Ann. Statist. 和Biometrika等SCI论文五十多篇) 。作为第一完成人研究成果获得教育部自然科学二等奖一次。

联系人：黄荣坦副教授

欢迎广大师生参加！