含有超越时间与相关收益率强度的极值风险度量  

Extreme Value Risk Measurement with the Excess Time and the Yield Strength

作　　者：陈守东[1,2] 周彻 

机构地区：[1]吉林大学数量经济研究中心,吉林长春130012 [2]吉林大学商学院,吉林长春130012

出　　处：《统计与信息论坛》2018年第1期3-10,共8页Journal of Statistics and Information

基　　金：国家社会科学基金重点项目<新常态下我国系统性区域性金融风险新特征及防范对策研究>(16AJY024);教育部人文社会科学重点研究基地重大项目<新常态下我国资本市场与经济增长的长期协调发展研究>(16JJD790016)

摘　　要：为研究不同市场状态下沪深300指数高频收益率的GARCH族期望波动率与极值风险之间的关系,采用非齐次的含有超越时间与相关收益率强度的二元极值方法分析系统性金融风险的VaR度量,结果表明:期望波动率的引入修正了齐次模型中GPD模型普遍退化的问题;预期波动率对尾部形状参数的正向影响在市场上升阶段比市场下降阶段更为明显,前一期期望波动率越高,则下一期单位期望波动率的变化对VaR的解释作用越小。In order to study the relationship between the expected volatility of the GARCH family and its extreme risk in the high market returns of the CSI 300 Index in different market conditions,the VaR measure of systematic financial risk is studied by using the binary extreme method with non-homogeneous rate of overtaking time and yield.The results show that the introduction of the expected volatility corrects the problem that the GPD model is generally degraded in the homogeneous model.The positive effect of the expected volatility on the tail shape parameter is more obvious in the market rising stage than the market decline stage,The higher the expected volatility of the previous period,the smaller the effect of the change of the expected volatility in the next period on VaR.

关 键 词：非齐次POT方法 期望波动率 VAR 极值 

分 类 号：O212.1[理学—概率论与数理统计] F224.9[理学—数学]