羊群效应检验以及与市场效率的关系研究

发布时间：2018-03-30 02:24

本文选题：股票市场　切入点：羊群效应　出处：《南京财经大学》2012年硕士论文

【摘要】：股票市场投资者非理性行为是行为金融研究的重要对象之一。投资者非理性行为不仅增加了股票市场的波动，还会影响股票市场效率。本文在对羊群效应相关理论、方法研究的基础上，探讨了羊群行为检验新方法，并对股票市场羊群行为检验的基础上，，进一步研究了市场羊群行为与市场效率的关系，主要内容如下: 一、股票市场羊群效应检验方法的改进及深圳股市羊群行为的检验（1）在综合CCK、I-CCK以及T-CCK模型的优点基础上，加速考虑市场的异方差效应，提出改进的羊群效应检验模型—V-CCK模型；（2）利用改进的V-CCK模型，对深圳股票市场羊群效应进行检验，并与已有模型检验结果比较，结果表明，提出的V-CCK模型更优；（3）静态结果表明，无论是股市上涨还是股市下跌，深圳股市都存在显著的羊群效应，但是下跌时的羊群效应强于上涨时的羊群效应；动态结果表明，随着时间的推移，该股市中的羊群效应呈现变弱的趋势，并且股市下跌时的羊群效应始终强于上涨时的羊群效应。二、深圳股票市场的羊群效应与市场效率的关系研究运用V-CCK模型和MF-DMA（多重分形移动平均消除趋势）方法分别提出羊群效应指标HE和有效性指标EFF，利用滑动窗技术，方法分别计算出羊群效应指标序列HEt和有效性指标序列EEFt。（1）相关性检验发现，HE和EFF之间存在明显的正相关关系；（2）线性Granger因果关系检验结果表明，HE不是EFF的线性Granger原因，而EFF是HE的线性Granger原因；（3）非线性检验表明，HE和EFF两个变量之间存在明显的非线性动态变化趋势，说明运用线性Granger因果检验得出的结论不准确。因此，本文利用VAR模型先过滤掉HE和EFF两者之间的线性关系，然后运用非线性Granger因果关系检验方法检验两者关系。结果表明，HE和EFF之间存在双向的非线性Granger因果关系，这表明：深圳股票市场羊群效应的强弱影响市场效率的高低，反过来，市场效率的高低也会影响羊群效应的强弱。
[Abstract]:The irrational behavior of investors in stock market is one of the important objects of behavioral finance. The irrational behavior of investors not only increases the volatility of stock market, but also affects the efficiency of stock market. Methods based on the research, the new method of herding behavior test is discussed, and the relationship between herding behavior and market efficiency is further studied on the basis of herding behavior test in stock market. The main contents are as follows:. The first is the improvement of herding effect test method in stock market and the test of herding behavior in Shenzhen stock market. 1) on the basis of synthesizing the advantages of CCK I-CCK and T-CCK model, considering the heteroscedasticity effect of market, an improved herding effect test model-V-CCK model is put forward. 2) using the improved V-CCK model, the herding effect of Shenzhen stock market is tested and compared with the existing model. The results show that the proposed V-CCK model is better; (3) the static results show that, whether the stock market is rising or falling, there is a remarkable herding effect in Shenzhen stock market, but the herding effect is stronger when the stock market falls than that in the rising stock market, and the dynamic results show that, as time goes on, the herding effect in Shenzhen stock market is stronger than that in the rising stock market. The herding effect in the stock market tends to be weaker, and the herd effect is always stronger when the stock market falls than when the stock market is rising. Second, the relationship between herding effect and market efficiency in Shenzhen stock market. Using V-CCK model and MF-DMA (multifractal moving average elimination trend) method, the herding effect index (HE) and the validity index (EFF) are proposed, respectively. Using sliding window technique, the herding effect index sequence HEt and the validity index sequence EEFtare calculated respectively. Correlation test showed that there was a significant positive correlation between HE and EFF. (2) the results of linear Granger causality test showed that HE was not the linear Granger cause of EFF, but EFF was the linear Granger cause of HE. 3) the nonlinear test shows that there is an obvious trend of nonlinear dynamic change between HE and EFF variables, which indicates that the conclusion obtained by using linear Granger causality test is not accurate. In this paper, VAR model is used to filter out the linear relationship between HE and EFF, and then the nonlinear Granger causality test method is used to test the relationship. The results show that there is a bidirectional nonlinear Granger causality between HE and EFF. This indicates that the herding effect in Shenzhen stock market affects the market efficiency, and in turn, the market efficiency will also affect the herding effect.
【学位授予单位】：南京财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F830.91;F224

【参考文献】