期望损失(ES)的几种逼近法比较  

Comparison of Several Approximation Methods of Expected Shortfall(ES)

作　　者：魏正元 李乔 王雪 郑小洋 WEI Zhengyuan;LI Qiao;WANG Xue;ZHENG Xiaoyang(Department of Statistics,Chongqing University of Technology,Chongqing400054,China)

机构地区：[1]重庆理工大学理学院,重庆400054

出　　处：《重庆理工大学学报（自然科学）》2021年第1期241-245,共5页Journal of Chongqing University of Technology：Natural Science

基　　金：重庆市自然科学基金项目(cstc2019jcyj-msxmX0386)。

摘　　要：基于Edgeworth和Saddlepoint展开,提出了对小样本情形下的ES计算问题的近似方法,给出了相应的逼近表达式,并将2种方法所得结果与基于Bootstrap方法的计算结果进行比较和分析。Monte Carlo模拟结果表明:Edgeworth展开法计算ES的值比Saddlepoint展开法更为精确。For the ES calculation problem with small samples,this paper proposes an approximation of the statistic based on Edgeworth expansion and Saddlepoint expansion,and gives the corresponding approximation expression,compares and analyzes the results obtained by the two methods with the results obtained by Bootstrap method.Monte Carlo simulation shows that Edgeworth method is more accurate than Saddlepoint method in calculating ES.

关 键 词：期望损失(ES) EDGEWORTH展开 Saddlepoint展开 不完全Gamma函数 

分 类 号：O212.1[理学—概率论与数理统计]