多元Log-PH分布的CTE风险度量  

Conditional Tail Expectations for Multivariate Log Phase Type Distributions

作　　者：戴泽兴 王传玉[1] 方颢[1] 

机构地区：[1]安徽工程大学数理学院,安徽芜湖241000

出　　处：《盐城工学院学报（自然科学版）》2014年第1期14-17,共4页Journal of Yancheng Institute of Technology:Natural Science Edition

基　　金：国家自然科学基金项目(612033139);安徽省重点教研项目基金(2012jyxm277)

摘　　要：CTE风险度量是一种关于重尾分布考虑了分位点以上部分平均损失的重要的度量方法。从一元Log-PH分布和多元PH分布的CTE风险度量研究,推广到多元Log-PH分布的CTE风险度量。结合文献[7]给出的次序统计量方法,运用多元Log-PH分布的Markov链性质,求解出多元Log-PH分布的CTE风险度量表达式。The conditional tail expectation is about heavy tail distribution for consider average loss above quantile. This article is an extension of the conditional tail expectation from Log phase - type distribution and multivariate phase - type distribution to multivariate Log phase - type distribution. Taking reference of order statistic method given in reference [ 7 ], using Markov chain properties of multivariate Log phase - type distribution, we construct the conditional tail expectation of multivariate Log phase - type distribution.

关 键 词：一元Log-PH分布 多元Log-PH分布 MARKOV链 CTE风险度量 

分 类 号：O226[理学—运筹学与控制论]