| 报告题目： | Automatic Detection of Significant Areas for Functional Data with Directional Error Control |
| 报 告 人： | 史建清 教授 |
| 英国 纽卡斯尔大学 | |
| 报告时间： | 周三（9月16日）下午3:30--4:30 |
| 报告地点： | 九龙湖校区教1-103 |
| 相关介绍： | To detect differences in the mean curves of two samples in longitudinal study or functional data analysis, we usually need to partition the temporal or spatial domain into several pre-determined sub-areas. The observations lying in each sub-area are assumed from the same population, and hypothesis test or multiple tests are used to detect the significance in each area. In my talk I will discuss a newly proposed method by using a large-scale multiple testing to find the significant sub-areas automatically in a general functional data analysis framework. A nonparametric Gaussian process regression model is introduced for simultaneous two-sided tests. The proposed procedure is asymptotically valid by controlling directional false discovery rates at any specified level. And it is computationally inexpensive and can accommodate different sampling schemes across the samples. I will also show some numerical examples including an application in an executive function research in children with Hemiplegic Cerebral Palsy. Jianqing Shi（史建清）教授2008年2月-3月曾任剑桥大学牛顿学院访问研究员，2012年8月至今为英国纽卡斯尔大学统计学教授，现任英国纽卡斯尔大学云机算和大数据研究培训中心副主任。曾担任国际统计界顶级学术期刊美国统计协会杂志Journal of the Royal Statistical Society: Series C 副主编（2010-2013）、英国EPSRC数学学科科研经费评审委员会委员（2011-2013）、英国APTS研究生课程全国执行委员会委员(2011-2012)、The Open Medical Informatics Journal的主编；先后被选为皇家统计学会（Royal Statistical Society）终身会员和泛华统计学会（International Chinese Statistical Association）终身会员。 |
