中国精算研究院

精算论坛讲座第184期—李政宵(4月26日)

发布时间:2021-04-21 10:25    浏览次数:[]

教育部人文社科重点研究基地中央财经大学中国精算研究院学术活动

精算论坛讲座第184

(2021426)

http://cias.cufe.edu.cn/__local/4/30/85/DB2C7339842D7DF119B95E6B08D_46BA1ED3_2634A.png

 

报告题目:A new class of copulas and its regressions for modelling multivariate heavy-tailed data

 

报告人李政宵 (对外经济贸易大学)

 

李政宵,中国人民大学经济学博士,现为对外经济贸易大学保险学院统计与精算系副教授,主要研究领域为精算统计模型,相依风险建模和巨灾风险管理。近年来在《ASTIN Bulletin: The Journal of the IAA》、《统计研究》、《数理统计与管理》、《系统工程理论与实践》、《保险研究》等国内外核心期刊发表多篇学术论文,出版学术专著一部,主持国家自然科学基金青年项目一项,先后参与国家社会科学基金重大项目、国家自然科学基金一般项目、教育部人文社会科学重点研究基地重大项目等多项课题研究。

 

摘要:

This talk proposes a new class of copulas that we call MGL copula. The new copula originates from extracting the dependence function of the multivariate generalized log-moyal-gamma (MGL) distribution whose marginals follow the univariate generalized log-moyal-gamma (GLMGA) distribution proposed by Li et al. (2021). The MGL copula can capture nonelliptical, exchangeable, and asymmetric dependencies among marginal coordinates and provide a simple formulation for regression applications.  We first discuss the probabilistic characteristics of MGL copula, and then obtain the corresponding extreme-value copula (MGL-EV copula). While the survival MGL copula can be also regarded as a special case of the MGB2 copula (Yang et al., 2011), we show that the proposed model is effective in regression modelling the dependence structures along with covariates. A simulation study is proposed to study the estimation method. Three applications are investigated to illustrate the usefulness of the proposed model.

 

报告时间:2021426 14:00--15:30

报告地点:腾讯会议(会议ID774 855 372

邀请人:池义春

欢迎各位老师和同学积极参加!