基于GARCH族模型的沪深300指数波动率预测  

Forecast of Volatility of Hu-Shen Index Based on GARCH

作　　者：严定琪[1] 李育锋[1] 

机构地区：[1]兰州大学数学与统计学院,甘肃兰州730000

出　　处：《兰州交通大学学报》2008年第1期92-95,共4页Journal of Lanzhou Jiaotong University

摘　　要：运用GARCH、EGARCH和GJR带正态分布和t分布的模型和方法对沪深300指数日收益率进行了统计拟合分析,得到了收益率序列尖峰厚尾性和异方差性等主要概率特征,并对GARCH、EGARCH、GJR带正态分布和t分布模型的预测效果进行了比较分析,发现基于学生t分布的GARCH(1,1)模型是最优的拟合模型,可以较好地提供沪深300指数未来两日的波动率预测.The rate of return of Hu-Shen 300 index is imitated and analysed by adopting GARCH, EGARCH and GJR with normal and student t-distribution model. The results show that the main probability characteristics of the rate of return of Hu-Shen 300 index are fat tails, excess kurtosis and heteroscedastic character. Furthermore, by comparing the predictive effect of GARCH,EGARCH and GJR with normal and student t-distribution model,we find that GARCH （1,1） with the student t-distribution model is the most superior fitting model which can give better forecast of volatility of Hu-Shen 300 index in two days.

关 键 词：沪深300指数 GARCH student-t分布 预测 

分 类 号：F224.0[经济管理—国民经济]