摘要
由于GARCH模型的系数固定不变,不能反映金融市场波动的结构变化,所以对于波动预测和动态风险管理都还不够完善。本文在GARCH模型中引入马尔科夫过程,从而使状态的转换体现在GARCH模型中,通过设置状态变量,构建马尔科夫状态转换GARCH模型(MRSGARCH),从而较好地揭示了存在结构转换的波动特性,对MRSGARCH模型进行参数估计,并给出了预测的详细过程,最后提出了MRSGARCH的波动持续性的估计方法。
The coefficients of GARCH are constant, and GARCH can't reflect the structural change of volatility, so GARCH is not perfect yet no matter for volatility forecast or dynamic risk management. The paper introduces Markov Regime switching GARCH model which allows the parameters of GARCH process to come from one of several different regimes, with transitions between regimes governed by an unobserved Markov chain. These changes to GARCH provide a better description of the volatility characteristics when structural change exists. We demonstrate the parameters estimation and forecast of MRSGARCH, and give out the calculation method of volatility persistence in the end.
出处
《数理统计与管理》
CSSCI
北大核心
2009年第4期637-645,共9页
Journal of Applied Statistics and Management
基金
国家自然科学基金项目(70471029)