本文系统地比较研究了中国和日本证券市场的多个问题,从数据挖掘和计量经济学角度分别针对价格序列、价格波动性和周内效应三方面对中日证券市场进行分析,得出指数收盘价时间序列比较方面,中国和日本两个证券市场的确存在一定的相似性,但中国市场的短期波动要大于日本市场;中国证券市场中上海和深圳股市的波动具有很强的波动聚类性和持续性,日本证券市场除了过去的方差记忆,还存在其他未知的影响市场变动的因素;周内效应在两国的体现不同等一系列重要结论。
This paper systematically studies some problems on security markets in China and Japan. From the aspects of data mining and econometrics, three main questions as price series, return violation and week-day effect have been studied both in Chinese and Japanese security markets. A series of important conclusions have been drawn as follows: Based on the method of time series data mining, the similarity of close price time series between main indexes in two countries' security markets is found, meanwhile, the short-term volatility is much stronger in China's market; China has a better 'volatility clustering and persistence than Japan's market; the week-day effect is different in two markets.