利用~Copula~的特点,灵活选择边缘分布模型、Copula~ 函数和时变参数演化方程,构建16个相关性模型.在此基础上,通过蒙特卡罗模拟,采用~VaR~和~ES~ 度量资产组合的市场风险,并通过回测检验比较不同模型的风险度量效果.以沪深~300~ 指数和恒生指数为样本构建投资组合进行实证研究,结果表明,边缘分布模型、Copula~时变参数演化方程和~Copula~ 函数的选择会影响风险度量的精度.在构建的~16~个相关性模型中,边缘分布为~MSM-EVT,时变参数演化方程为~GAS~模型, Copula~函数为 Rotated Gumbel Copula~的~MSM-EVT-R-GAS~模型风险度量效果最好.

Taking advantage of Copula, this paper chose marginal model,Copula function and the evolution equation of the time varying parameter for Copula with flexibility,then constructed 16 correlation models. Base on these correlation models, we used VaR and ES to measure the market risk of the portfolio through the technique of monte carlo simulation. Furthermore, we used backtesting test to compare the performance of the risk measurement of these different models. We selected daily data of CSI 300 Index and the Heng Seng Index as samples to conduct empirical study. The empirical results showed that the marginal model, evolution equation of the time varying parameter for Copula and the choose of Copula will affect the accuracy of risk measurement. Among the 16 correlation models, MSM-EVT-R-GAS model can measure the risk accurately.