摘要
在金融投资组合的风险管理中,估计极端事件的概率,是一个重要的问题.阐述了极值理论和两类极值分布,以上证综合指数为例,将极值理论用于风险价值的计算,给出了VaR和ES的估计值,并与传统方法得出的结果进行了比较分析,结论表明,用极值方法度量金融风险具有很高的准确性.
Assessing the probability of rare and extreme events is an important issue in the risk management of financial portfolios. This paper expatiates Extreme Value Theory（EVT） and two kinds of distributions . Taking Composite Stock Index of Shanghai Stock Exchange Centre for example ,the article applies EVT to compute value at risk and gives the estimated value of VaR and ES. Then,the estimation results are compared to conventional methods. The conclusion shows that the EVT methods have good veracity in estimating financial risk.
基金
安徽省教育厅自然科学基金资助项目(2003kj036)
关键词
风险价值
极值理论
广义极值分布
广义帕雷托分布
Value-at-Risk
Extreme Value Theory
Generalized Extreme Value Distribution
Generalized Pareto Distribution
