论文部分内容阅读
本文使用我国沪深股市综合指数日收盘价数据,运用ARFIMA模型、FIGARCH模型以及ARFIMA-FIGARCH模型对我国股票收益率序列和均值回复性与长期记忆性进行了检验。检验结果表明,我国沪深股市收益率序列中存在微弱的长期记忆性特征,但波动率序列中存在非常显著且较强的长期记忆性效应。同时,我们利用Student-t分布来刻画我国沪深股市收益率序列的尾部特征,发现该序列中存在明显的“尖峰厚尾”分布性质,也在TGARCH模型中发现了股票收益率波动中对“利空消息”的非对称效应。
In this paper, we use the daily closing price data of Shanghai and Shenzhen Stock Market Composite Index, using ARFIMA model, FIGARCH model and ARFIMA-FIGARCH model to test China’s stock return series and mean recovery and long-term memory. The test results show that there is a weak long-term memory characteristic in the yield series of Shanghai and Shenzhen stock markets, but there is a very significant and strong long-term memory effect in the volatility series. At the same time, we use the Student-t distribution to characterize the trailing features of the yield series in Shanghai and Shenzhen stock markets. We find that there are obvious “thick tail” distributions in this series. We also found the volatility of stock returns in the TGARCH model Asymmetric effect on “bad news”.