Generalising Dynamic Semiparametric Averaging Forecasting for Time Series with Discrete-valued Response

发布时间：2023-07-13 作者：浏览次数：

Speaker:	卢祖帝	DateTime:	2023年7月20日（周四）上午10:30-11：30
Brief Introduction to Speaker: 卢祖帝，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的统计学终身讲席教授(Chair in Statistics)、博导，主要研究兴趣为非线性时间序列分析、金融统计、计量经济学、非线性时空大数据分析等。他是国际上非线性时空间数据统计学的主要研究者和倡导者之一。卢教授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(Australian Future Fellow, 相应于中国国家杰出青年基金)和欧盟居里夫人研究基金项目(Career Integration Grant/Marie Curie Fellow)及多项面上项目的资助，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, Journal of Royal Statistical Society series B, Journal of Econometrics, Econometric Theory等发表90多篇学术论文。同时他在国内外杂志Journal of Time Series Analysis, Environmental Modelling and Assessment, Cogent Research in Mathematics and Statistics（负责统计版块）、《系统工程理论与实践》等担任副主编、高级编辑和编委。		Place:	6号楼M201报告厅
Abstract:In this paper, we propose to explore how to utilise the useful high-dimensional dynamic lagged information for forecasting of time series data with discrete-valued response. Our approach will generalise the existing flexible semiparametric marginal regression model averaging (MARMA) forecasting of Li, Linton and Lu (2015), which has been shown a useful data-driven method, but was designed for nonlinear forecasting of continuous valued time series by a least squares averaging. We have hence suggested a generalised MARMA (GMARMA) procedure under a general time series conditional exponential family of distributions, which flexibly accommodates nonlinear forecasting of discrete-valued response, and further allowing the lagged effects including discrete-valued information for forecasting. A conditional likelihood model averaging method, instead of the least squares, is thus developed for the averaging weights estimation in the GMARMA, under beta-mixing time series data generating process...

上一条： Recent progress on the positivities in several complex variables and complex geometry
下一条： Joint Analysis of Mixed Types of Outcomes with Latent Variables