2019概率统计及其应用系列报告之五(吴月华教授)

发布时间:2019-04-24作者:访问量:68

报告题目: Multiple change-points detection in generalized linear models

报 告 人: 吴月华 教授

报告时间: 2019425(周四)  16:00-17:00

报告地点: 磬苑校区数学科学学院H306

报告摘要:In this talk, we focus on the problem of multiple change points estimation in GLMs in which both number of change points and their locations are unknown. We propose a simultaneous multiple change points estimation method which first partitions the data sequence into several segments to construct a new design matrix, secondly convert the multiple change points estimation problem into a variable selection problem, and then estimate the regression coefficients by maximizing a penalized likelihood function. The consistency of the coefficient estimator is established in which the number of coefficients can diverge as the sample size goes to infinity. The nonzero coefficient estimates provide the information about which segments potentially contain a change point. An algorithm is provided to estimate the multiple change points. Simulation studies are conducted for both logistic and log-linear models. A real data application is also presented. Joint work with X. Sun

欢迎各位老师、同学届时前往!

 

                                              科学技术处

                                            2019424


返回原图
/