论文部分内容阅读
利用GARCH、EGARCH和GJR带正态分布和t分布的GARCH模型族对沪深300指数日收益率进行了统计拟合比较分析,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和GJR带正态分布和t分布的GARCH模型族的预测效率进行了比较分析,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。
Using the GARCH model family with normal distribution and t-distribution of GARCH, EGARCH and GJR, we compare the daily returns of Shanghai and Shenzhen 300 index by statistical fitting, and get the main probabilistic characteristics of peak-tail and heteroscedasticity of yield series , And compared the GARCH, EGARCH and GJR forecasting efficiencies of GARCH model family with normal distribution and t distribution. It is found that the GARCH (1,1) model based on student t distribution is the best fitting model, which can compare Provide good CSI300 index volatility forecast for the next two days.