报告时间：2019年12月24日（星期二）10:00

报告地点：翡翠湖校区翡翠科教楼B座1710

报 告 人：卢祖帝 教授

工作单位：英国南安普顿大学

举办单位：数学学院

报告人简介：

卢祖帝，1996年获中国科学院博士学位，现为英国南安普顿大学数学科学学院和南安普顿统计科学研究所统计学讲席教授、博士生导师。研究领域为金融统计、计量经济学、非线性时间序列分析和非线性时空大数据和智能建模等。卢祖帝教授先后任职于中国科学院数学与系统科学研究院、比利时鲁汶天主教大学、英国伦敦经济学院、澳大利亚科廷大学和阿德莱德大学。曾先后获得中国国家自然科学重点基金、澳大利亚国家研究理事会未来研究杰出青年基金项目(ARC Future Fellow)和欧盟居里夫人研究基金项目(Career Integration Grant/Marie Curie Fellow)等多项各种面上项目的支助，是国际统计学会的当选会员(Elected Member)。已在国际统计学和计量经济学的主要杂志包括顶级期刊Annals of Statistics, Journal of American Statistician Association,Journ-al of Royal Statistical Society Series B, Journal of Econometrics, Econometric Theory等发表80多篇学术论文。担任Journal of Time Series Analysis的副主编 及Environmental Modelling and Assessment 和 Cogent Mathematics and Statistics等国际期刊的编委。

报告简介：

In many applications related to spatial problems, to study possibly nonlinear relationship betw-een covariates and the concerned response at a location, accounting for the temporal lag intera-ctions of the response at a givenlocation and the spatio‐temporal lag interactions between locati-ons could improve the accuracy of estimation and forecasting. There lacks, however, methodol-ogy to objectively identify and estimate such spatio‐temporal lag interactions. In this talk, we pr-esent a semiparametric data‐driven nonlinear time series regression method that accounts for l-ag interactions across space and over time. A penalized procedure utilizing adaptive Lasso is dev-eloped for the identification and estimation of important spatio‐temporal lag interactions. The-oretical properties for our proposed methodology are established under a general near epoch d-ependence structure and thus the results can be applied to a variety of linear and nonlinear tim-e series processes. For illustration, we analyze the US housing price data and demonstrate subs-tantial improvement in forecasting via the identification of nonlinear relationship between HPI and CPI as well as spatio‐temporal lag interactions.