题目：尾部风险度量Expectile及其相关研究
报告人：张奕（浙江大学）
时间：2018年4月16日（周一）下午13：30
地点：信息楼407房间
报告摘要：
We construct a specific form of piecewise distortion function which can distort a random risk to its expectile and define a class of expectile distortion measure which is generated from the piecewise form distortion. The consistent estimation of the expectile distortion parameter is given by the maximum empirical likelihood method. As the application, we discuss a new premium principle based on the expectile distortion measure. We propose a new method for estimating conditional Value-at-risk (VaR) and expected shortfall (ES) of financial time series, which based on the combination of composite expectile and empirical likeli-hood. The method of composite expectile is used to estimate the conditional variance of conditional heteroscedasticity model such as GARCH.

报告人简介：
张奕，浙江大学数学科学学院、统计研究所教授、博士生导师。研究领域包括精算学、统计大样本理论、计量经济学等, 主要关注重尾分布下的随机和的极限问题、相依风险度量问题、最优再保险问题，长寿风险度量和管理问题、精算中的统计推断问题。近来对风险度量，特别是极端事件的风险度量有丰富的研究经验。近十年内已在《Insurance：Mathematics & Economics》、《Scandinavian Actuarial Journal 》、《Stochastic Processes& their Applications》、《Journal of Applied Probability》、《Statistics & Probability Letter》、《Journal of Nonparametric Statistics》、《Applied Stochastic Models in Business and Industry》、《Journal of Multivariate Analysis》、《Methodology and Computing in Applied Probability》等重要国际精算学和概率统计学术期刊发表论文30余篇，其中SCI收录20余篇；承担国家级、省部级课题10余项，其中主持国家自然科学面上项目、国家社科基金重大项目子项目、教育部人文社科项目各一项、浙江省自然科学基金两项；承担企事业单位的横向研究课题若干。