中国股市波动过程长期记忆性影响因素研究

An Empirical Research on the Influencing Factors of Long Memory of Volatility Process in Chinese Stock Market

以ICSS方法为基础,使用参数和半参数估计方法研究了上证综合指数和深圳成分指数波动过程的变结构与长期记忆性的相关关系,对中国股市日间收益波动方差序列的长期记忆性进行了估计。研究结果发现中国股市日间收益波动的方差序列存在显著的长期记忆性,并且在剔除了变结构的因素之后,其波动无断点的方差序列仍然存在长期记忆性,但程度有所减弱。研究结果表明,未被考虑的体制变化是长期记忆性的影响因素,但不能解释全部的长期记忆性。

Based on the ICSS method, the existence of long memory and structural break is tested in Chinese stock market for Shanghai synthesized index and Shenzhen composite index. Both parametric and semi-parametric methods are used to estimate the long memory parameter for variance process of daily returns. While long memory is evident in the variance processes, there is some evidence of long memory for break-free series, although the degree of persistence is weaker. A structural break analysis reveals that this feature is partially explained by unaccounted changes in regime.