学术报告
Heteroscedasticity Identification and Variable Selection via Multiple Quantiles Regression-王明秋（曲阜师范大学）
题目：Heteroscedasticity Identification and Variable Selection via Multiple Quantiles Regression
报告人：王明秋（曲阜师范大学）
Abstract :
In general, high-dimensional data often display heterogeneity. If the heteroscedasticity is neglected in the regression model, then it will produce inefficient inferences of the regression coefficients. Quantile regression is robust to outliers and can also accommodate the heterogeneity, so we apply multiple quantiles regression to identify the heteroscedasticity, common features of quantile coefficients and eliminate irrelevant variables simultaneously based on the regularized approach. We also give the theoretical properties of the proposed method under some regularity conditions. Some simulation studies are conducted to illustrate the finite sample performance of the proposed method. Our method can identify the covariates that affect the variance of the response. Boston house data is analyzed to support our new method.
个人简介：
王明秋，副教授，硕士生导师。现为统计学院应用统计系主任，中国现场统计研究会数据科学与人工智能分会理事。研究兴趣包括稳健估计、非参数统计推断、变量选择、高维数据分析等。先后多次前往香港大学、南方科技大学进行学术访问。主持或参与国家自然科学基金4项、省部级项目3项。在国内外学术刊物上Journal of Statistical Planning and Inference、Journal of Nonparametric Statistics、Acta Mathematica Sinica-English Series等发表论文30余篇。先后被评为曲阜师范大学师德先进个人、曲阜师范大学优秀教师。
时间：2020年10月29日（周四）20：00-21：00
地点：线上腾讯会议（会议号：766 943 031）
联系人：周洁，胡涛
