Dynamic risk measurement of bivariate portfolio and the selection of copula: Evidence from the American Stock Indices
Copula theory has been extensively applied to estimate the risk of a portfolio, and the selection of appropriate copula functions is an important issue of the risk estimation In this paper, the change of an individual asset in the portfolio is modeled by GARCH, and three common copulas Normal, Students t, and Clayton copula are used to estimate the VaR of the bivariate portfolio composed of SP500 and Dow Jones Index, and back-testing method are employed to analyze the estimation results It is founded that violation rate and mean of absolute errors can be used as criteria to choose copula functions based on an investors attitude toward the risk.
copula selection portfolio Value at Risk back-testing
Xiaoping Zhou Guoxiao Yang
Department of Mathematics,Beijing Institute of Technology,Beijing 100081,China
国际会议
2008 International Conference on Risk and Relianility Management(2008风险与可靠性管理国际会议)
北京
英文
160-164
2008-11-10（万方平台首次上网日期，不代表论文的发表时间）