机构地区: 天津大学理学院
出 处: 《数量经济技术经济研究》 2006年第7期126-133,共8页
摘 要: 为了反映金融时间序列的波动集聚性、非对称性、厚尾性以及在实证研究中表现出的伪持续性,本文结合门限GARCH模型以及变结构的方法提出了变结构门限t-GARCH模型。首先用MonteCarlo模拟的方法考虑了变结构GARCH模型中存在的伪持续性问题;其次针对金融时间序列非对称性、厚尾性以及强持续性的特点提出了变结构门限t-GARCH模型,总结了关于变结构点检验的几个主要方法;最后用该模型来拟合沪市和深市两个股市的周收益率序列,得到了比GARCH模型更好的拟合结果。 The threshold t - GARCH model with structural change is proposed to depict the asymmetry, tail - fatness and strong persistence of financial time series. Spurious persistence of GARCH model with structural change is considered by Monte Carlo simulation, then threshold t - GARCH model with structural change is defined and some methods of structural change test are introduced. Empirical results on weekly return series of Shanghai and Shenzhen stock markets indicate that threshold t - GARCH model with structural change outperforms the GARCH model in capturing the heteroskedasticity , the asymmetry and tail - fatness, and avoiding the potential spurious volatility persistence.
关 键 词: 变结构门限 模型 变结构点 伪持续性 分布 模拟
领 域: [经济管理]