报告人：陈孝伟(南开大学)

时间：2023年12月21日 15:00-

地址：数统学院LD402

摘要：This article presents a high-dimensional quantile regression framework to forecast Value at Risk (VaR) and Expected Shortfall (ES) simultaneously based on the nature of Multivariate Asymmetric Laplace (MAL), where "high-dimension" implies a large bunch of assets and associated idiosyncratic factors. We generalize the traditional joint quantile regression framework to a time-varying setting, which allows us to capture the dynamic tail interdependence among assets. The proposed methodology is highly flexible and computationally tractable. A closed-form likelihood expression allows for straightforward parameter estimation, and the factor structure makes the model scalable to high dimensions. Applying this new statistical model to a panel of 50 stocks from 11 sectors of S&P 500 index over 2001-2021, we show that our model performs well on out-of-sample VaR and ES.

简介：南开大学金融学院精算学系教授、博士生导师，美国伊利诺伊大学、韩国木浦海洋大学访问学者，研究方向为保险精算学、定量风险管理，发表论文50余篇。主持参与国家级、省部级和横向课题十几项。入选天津市首批“用三年时间引进千名以上高层次人才”项目、天津市131第三层次人才、获得中国运筹学会不确定系统分会“钟家庆运筹学奖”。担任中国运筹学会不确定系统分会理事长、中国运筹学会理事，ABS3*国际期刊《Fuzzy Optimization and Decision Making》副主编。

邀请人：张志民

欢迎广大师生积极参与！