基于GARCH-EVT-Copula的社保基金投资组合风险测度研究  

On the Risk Measure of the Social Security Fund Investment Portfolio based on GARCH-EVT-Copula

作　　者：江红莉[1] 何建敏[1] 庄亚明[1] 

机构地区：[1]东南大学经管学院,江苏南京211189

出　　处：《金融理论与实践》2011年第8期8-12,共5页Financial Theory and Practice

基　　金：国家自然科学基金项目(70671025/G0115)

摘　　要：社保基金是社会保障事业健康发展的物质基础,安全性是其投资的首要原则。文章基于GARCH-EVT-Copula方法测度了社保基金投资组合的VaR。首先,基于GARCH、EVT对投资组合中各金融资产收益的边缘分布建模,然后,采用极大似然估计法和Bootstrap方法估计尾部的分布函数,接着,基于Copula方法研究组合中金融资产间的相关结构,最后,运用Monte Carlo方法测度投资组合的VaR。Kupiec检验表明,基于GARCH-EVT-Copula模型测度社保基金投资组合的风险是合适的。Social security fund is the material base for healthy development of social security business, which gets the security to be the first principle for investment. This paper measured VaR of portfolio in social security fund with the GARCH-EVT-Copula approach. Firstly, distribution modeling by using GARCH and EVT based on the brink of all financial assets in portfolio, then figured out the tails on the distribution functions by using Maximum Likehood Estimation （MLE） and Bootstrap, and then research- ing on relational structure in financial assets portfolio based on Copula approach, finally the measure- ment was using the Monte Carlo method for VaR of portfolio. The inspection of Kupiec showed that GARCH-EVT-Copula model is suitable for measurement of risks of social security fund portfolio.

关 键 词：GARCH 极值理论 Coupla 社保基金 投资组合 自助法 

分 类 号：F840.32[经济管理—保险]