学术报告
报告题目：Robust Mixture Regression and Outlier
Detection via Penalized Likelihood
报 告 人：姚卫鑫 教授
Department of Statistics，University of California, Riverside
时 间：2015年06月26日(星期五)上午10:00
地 点：数学楼第四报告厅
摘 要：Finite mixture regression models have been widely used for modelling mixedregression relationships arising from a clustered and thus heterogenous population. The classical normal mixture model, despite of its simplicity and wideapplicability, may fail dramatically in the presence of severe outliers. We propose a robust mixture regression approach based on a sparse, case-specic, andscale-dependent mean-shift parameterization, for simultaneously conducting outlier detection and robust parameterestimation. A penalized likelihood approachis adopted to induce sparsity among the mean-shift parameters so that the outliers are distinguished from the good observations, and a thresholding-embeddedExpectation-Maximization (EM) algorithm is developed to enable stable and efficient computation. The proposed penalized estimation approach is shown to havestrong connections with other robust methods including the trimmed likelihoodand the M-estimation methods. Comparing with several existing methods, theproposed methods show outstanding performance in numerical studies.
简 历：Dr. WeixinYao is anassociate professor in the Department of Statistics, University of California, Riverside. He received PhD degree in the Pennsylvania State University-University Park in 2007. His research interests cover Mixture models, Analysis of longitudinal data, High dimensional data analysis, Nonparametric andsemiparametric modeling, Robust statistics, Variable selections. His honors and awards include the William L. Harkness Graduate Teaching Award in 2006 and Vollmer-Kleckner Fellowship in 2004.In last several years, he has had many publications.For example, he writed Minimum profile Hellinger Distance Estimation for a Semiparametric Mixture Model for The Canadian Journal of Statistics. He also has other articles which were published in Scandinavian Journal of Statistics, Journal of Business and Economics Statistics, Computational Statistics and Data Analysis, Journal of American Statistical Association and Journal of Nonparametric Statistics. He is also the president of Kansas-Western Missouri Chapter of the American Statistical Association (2013-2014) and vice president of Kansas-Western Missouri Chapter of the American Statistical Association
(2012-2013).
欢迎广大师生参加！
统计研究院
2015年6月17日