基于连续信念系统的中国A股市场不对称羊群行为研究

发布时间：2018-01-18 00:10

本文关键词：基于连续信念系统的中国A股市场不对称羊群行为研究　出处：《华中科技大学》2012年硕士论文　论文类型：学位论文

【摘要】：基于扩展的连续信念系统(CBS)，本文对中国A股市场不对称羊群行为的存在性及其对波动产生的不对称效应进行了研究。采用的数据是从1991年7月3日到2012年7月31日的沪深两市A股流通市值加权市场指数，数据来源于CCER数据库。并根据股市规范程度和经济背景的不同，将整个时期分为不规范的初始时期，缓慢震荡上升的中间时期和全球金融危机爆发后的危机时期三个子时期，对这三个时期进行了比较分析。根据扩展的CBS的假设，经过推导发现不对称羊群行为可以用一个不对称GARCH（1,1）模型来模拟。不对称GARCH（1,1）模型的估计表明：三个子时期都存在不对称羊群行为；但是，在中间和危机时期，负向冲击下的羊群行为比正向冲击下的严重，这与我们通常看到的现象相符合，但初始时期刚好与后两个时候相反，即正向冲击下的羊群行为比负向冲击下的严重，我们称之为反不对称羊群行为；还有危机时期是羊群行为最不对称的时期。随后，本文用EGARCH（1,1）模型研究分析了收益率波动的不对称，发现各个时期收益率波动的不对称特征与羊群行为的不对称特征表现的完全一致。不对称羊群行为和波动之间的这种清楚的联系强有力地支持了不对称羊群行为效应假说，也说明可以用不对称羊群行为来解释波动的不对称性，尤其是传统假说（比如杠杆效应，反馈效应）解释不了的反不对称性。这个结论提示可以通过防止不对称羊群行为的发生来减少中国A股市场收益率的不对称波动，稳定中国的经济发展。
[Abstract]:An extended continuous belief system (CBS). This paper studies the existence of asymmetric herding behavior and its asymmetric effect on volatility in Chinese A-share market. The data used are from July 3rd 1991 to July 31st 2012 in Shanghai and Shenzhen. Shanghai A-share circulation market value weighted market index. The data come from CCER database, and the whole period is divided into the non-standard initial period according to the stock market normative degree and the different economic background. In the middle period of slow shock rise and the three sub-periods of crisis period after the global financial crisis, the three periods are compared and analyzed. According to the extended CBS hypothesis, this paper makes a comparative analysis of these three periods. It is found that asymmetric herding behavior can be simulated by an asymmetric GARCH(1 model. 1) the estimation of the model shows that there is asymmetric herding behavior in all three sub-periods; However, in the middle and crisis period, the herding behavior under the negative shock is more serious than that under the positive shock, which is consistent with the phenomenon we usually see, but the initial period is exactly the opposite of the last two times. That is, the herd behavior under positive shock is more serious than that under negative shock, which is called anti-asymmetric herding behavior. There is also a crisis period is the most asymmetric period of herding behavior. Then, this paper uses EGARCH1) model to analyze the asymmetry of return volatility. It is found that the asymmetric characteristics of return volatility in each period are completely consistent with the asymmetric characteristics of herding behavior. This clear link between asymmetric herding behavior and volatility strongly supports asymmetric herding behavior. Effect hypothesis. It also shows that asymmetric herding behavior can be used to explain the asymmetry of volatility, especially the traditional hypothesis (such as leverage effect). This conclusion suggests that the asymmetric herding behavior can be prevented to reduce the asymmetric volatility of the yield in China's A-share market and stabilize China's economic development.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【参考文献】