Modelling function-valued processes with non-separable and/or non-stationary covariance structure
报告人：Prof. Jianqing Shi (Department of Statistics and Data Science, Southern University of Science and Technology, Shenzhen, China\ and Alan Turing Institute, UK )
时间：2020-12-10 16:00-17:00
地点：Room 1114, Sciences Building No. 1
Abstract: Separability of the covariance structure is a common assumption for function-valued processes defined on two- or higher-dimensional domains. This assumption is often made to obtain an interpretable model or due to difficulties in modelling a potentially complex covariance structure, especially in the case of sparse designs. We proposed to use Gaussian processes with flexible parametric covariance kernels which allow interactions between the inputs in the covariance structure. When we use suitable covariance kernels, the leading eigen-surfaces of the covariance operator can explain well the main modes of variation in the functional data, including the interactions between the inputs. The results are demonstrated by simulation studies and by applications to real world data.
报告人简介: 史建清，南方科技大学统计与数据科学系教授，理学院生物医学统计中心主任，英国国家艾伦图灵研究院图灵研究员，英国皇家统计学会会士。曾任英国纽卡斯尔大学（Newcastle University）统计学教授。主要研究方向包括函数型数据分析，生物医学统计，缺失数据分析，meta-analysis等。在国际学术刊物上发表高水平学术论文多篇，包括统计顶级期刊 JRSSB, JASA, Biometrika 和Biostatistics。曾任英国皇家统计协会《应用统计》副主编，Guest AE for JRSS discussion paper，英国纽卡斯尔大学云计算和大数据研究培训中心副主任。曾获邀任剑桥大学世界最顶级数学学院之一牛顿学院访问研究员，曾获IEEE康复游戏和健康国际年会最佳论文奖、美国统计协会非参数统计分会年度最佳论文奖。2011年在Chapman & Hall 出版专著：Gaussian Process Regression Analysis for Functional Data。
