VaR is subject to a significant positive bias

被引:18
作者
Inui, K
Kijima, M [1 ]
Kitano, A
机构
[1] Kyoto Univ, Grad Sch Econ, Sakyo Ku, Kyoto 6068501, Japan
[2] Financial Serv Agcy, Chiyoda Ku, Tokyo 1008967, Japan
关键词
value-at-risk; Harrell-Davis estimator; historical simulation; concave ordering;
D O I
10.1016/j.spl.2005.02.001
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
This article shows that value-at-risk (VaR), the most popular risk measure in financial practice, has a considerable positive bias when used for a portfolio with fat-tail distribution. The bias increases with higher confidence level, heavier tails, and smaller sample size. Also, the Harrell-Davis quantile estimator and its simulation counterpart, called the bootstrap estimator, tend to have a more significant positive bias for fattail distributions. (c) 2005 Elsevier B.V. All rights reserved.
引用
收藏
页码:299 / 311
页数:13
相关论文
共 14 条
[1]  
[Anonymous], 1970, CONTINOUS UNIVARIATE
[2]   Coherent measures of risk [J].
Artzner, P ;
Delbaen, F ;
Eber, JM ;
Heath, D .
MATHEMATICAL FINANCE, 1999, 9 (03) :203-228
[3]   How accurate are value-at-risk models at commercial banks? [J].
Berkowitz, J ;
O'Brien, J .
JOURNAL OF FINANCE, 2002, 57 (03) :1093-1111
[4]  
DIELMAN T, 1994, COMM STAT SIMULATION, V23, P203
[5]  
HARRELL FE, 1982, BIOMETRIKA, V69, P635
[6]  
Heikkinen V.P., 2002, J. Risk, V4, P77, DOI [10.21314/JOR.2002.065, DOI 10.21314/JOR.2002.065]
[7]  
HENDRICKS D, 1996, FEDERAL RESERVE APR, P39
[8]   On the significance of expected shortfall as a coherent risk measure [J].
Inui, K ;
Kijima, M .
JOURNAL OF BANKING & FINANCE, 2005, 29 (04) :853-864
[9]  
Jorion P, 2000, Value at Risk
[10]  
Kijima M., 1996, Mathematical Finance, V6, P237, DOI DOI 10.1111/MAFI.1996.6.ISSUE-3