主  讲  人：沈俊山 博士 （北京大学数学科学学院）
时      间：2015年11月5日 10:30—11:30
地      点：必赢76net线路官网北一区文科楼 707 教室
主 办 单 位：必赢76net线路官网数学科学学院
内 容 介 绍：In this paper, we propose a mean-covariance regression model for clustered data. A mixture model is used to describe the non-normal distribution of the regression errors.  An adjusted Dirichlet process is constructed with the empirical likelihood method and it  can be used as a prior for  mixing distributions with given mean-covariance structure.  Two simulation studies are conducted to evaluate the performance of the proposed method.  Finally, the proposed method is applied to analyze the data from a sleep deprivation study.