摘要
针对金融资产收益的异常变化,采用SV-GED模型捕捉收益分布的厚尾性、波动的异方差性等特征,将收益序列转化为标准残差序列,结合SV-GED模型与极值理论拟合标准残差的尾部分布,建立了一种新的度量金融风险的基于EVT-POT-SV-GED的动态、VaR模型.用该模型对上证综指做实证分析,结果表明该模型能够更精确、合理地度量上证综指收益的风险.
Considering the characteristic of thick tail and fluctuation heteroscedasticity of financial assets, the paper uses SV-GED models together with the extreme value theory to establish a new financial risk measure model-the dynamic VaR model based on EVT-POT-SV-GED. Shanghai Stock Exchange（SSE） composite in- deides are used for our empirical analysis, and the results show that the dynamic VaR model, which combines the SV-GED model and the extreme value theory, is more rational and effective in measureing stock earnings risk and it is more advantageous than several other models in fitting the tail distribution of standard residual income and measuring SSE composite index earnings risk.
出处
《系统工程学报》
CSCD
北大核心
2012年第2期152-159,共8页
Journal of Systems Engineering
基金
国家自然科学基金资助项目(70473107)
