主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2006, Vol. ›› Issue (5): 1-6.

• 论文 •    下一篇

金融资产的VaR和CVaR风险的优良估计

刘小茂, 杜红军   

  1. 华中科技大学主校区数学系, 武汉, 430074
  • 收稿日期:2005-12-01 修回日期:2006-09-03 出版日期:2006-10-28 发布日期:2012-03-07
  • 基金资助:
    广西科学研究与技术开发资助项目(0385008)

Optimal Estimation of Value-at-Risk and Conditional Value-at-Risk

LIU Xiao-mao, DU Hong-jun   

  1. Mathematics Department, Huazhong University of Science and Technology, Wuhan 430074, China
  • Received:2005-12-01 Revised:2006-09-03 Online:2006-10-28 Published:2012-03-07

摘要: 本文利用统计理论的优良点估计方法来估计金融市场风险的VaR和CVaR,既可避开现有方法中大量的模拟计算和参数估计等工作,又可提高估算精度.在资产-正态模型下,根据不同的风险估计要求,对金融资产的这两种风险分别提供了三种优良估计,即一致最小方差无偏估计,最佳线性次序统计量无偏估计,最佳线性次序统计量同变估计,并提供了实证分析.

关键词: 风险价值, 条件风险价值, 一致最小方差无偏估计, 最佳线性次序统计量无偏估计, 最佳线性次序统计量同变估计

Abstract: In this paper,statistical method is used to improve the estimation of value-at-risk(VaR) and conditional value-at-risk(CVaR).These methods can avoid burdensome simulation calculation or parameters estimation and improve estimation precision.This paper discusses the optimal estimation of value-at-risk and conditional value-at-risk for assets under normal distribution and gives the uniformly minimum variance unbiased estimates(UMVUE),the best linear unbiased estimates(BLUE) and the best linear invariant estimates(BLIE) of VaR and CVaR based on order statistics.Furthermore,we show the practicability and validity of these methods through empirical analysis.

Key words: VaR, CVaR, the uniformly minimum variance unbiased estimate, the best linear order statistics unbiased estimate, the best linear order statistics invariant estimate

中图分类号: