论文部分内容阅读
假设风险满足一定条件,利用Copula和蒙特卡洛模拟方法对边际分布的选择和边际分布间的尾部相关关系两个方面进行分析,并得出结论:第一,在一定条件下,边际分布的选择并不影响对总风险的估计;第二,相关系数模型可能因为边际分布间的尾部相关关系而低估总风险.在此基础上,提出了尾部相关系数的概念,得出了修正的相关系数模型.
Assuming that the risk satisfies certain conditions, we use Copula and Monte Carlo simulation methods to analyze the marginal distribution and the tail correlation between the marginal distributions and draw the conclusion: First, under certain conditions, the choice of marginal distribution Second, the correlation coefficient model may underestimate the total risk because of the tail correlation between marginal distributions.On this basis, the concept of tail correlation coefficient is proposed, and the modified correlation coefficient model .