摘要
为更好地理解我国股票市场运行状态,使用局部Whittle(local Whittle,LW)半参数法分别估计了沪/深股市在牛/熊市阶段波动率的长记忆性强度.实证结果显示:沪(深)市波动率在牛/熊市阶段均存在显著的长记忆性,且熊市波动的长记忆强度要大于牛市;在牛(熊)市阶段,沪市波动的长记忆强度大于深市,且两者的差异在牛市阶段更为明显.特别的,对波动序列作了长记忆真伪性检验,增强了估计和解读的合理性.
To get a better understanding of the operational status China's stock market, the local Whittle semi-parametric method was used to estimate the long memory in bull and bear market volatility. The empirical results show that stock market volatility in each market phase present significant long memory, and that the volatility was more persistent in the bear market than in the bull market. Comparison between Shanghai and Shenzhen exchanges in the same market phase found that the volatility was more persistent in Shanghai than in Shenzhen, and that the difference was more obvious in the bull market. In particular, the rationality of our inquiry was enhanced by performing a test that can distinguish between true and spurious long memory.
关键词
波动率
长记忆性
半参数
LW估计
结构变化
volatility
long memory
semiparametric
local Whittle estimator
structural change