从大股东联结网络视角探讨股价联动关系

发布时间：2018-03-07 01:27

本文选题：股东联结网络　切入点：联动　出处：《西南交通大学》2013年硕士论文　论文类型：学位论文

【摘要】：股票价格相关性对于风险管理、投资决策具有重要影响。对于股价相关性的研究,现代金融理论主要基于经济基本面进行解释,即认为股价变动的相关性来源于影响资产现金流和影响资产折现率的基本面因素。已有研究表明,证券间相关程度远超出了经济基本面因素的影响,这种过度联动对金融理论与实践具有重要影响而受到广泛关注。现有研究认为机构投资者的市场分割是导致过度联动的重要因素,但市场分割如何导致过度联动的微观机制尚缺乏足够的认识。本文从信息扩散和关联交易角度讨论投资者对股价联动的影响,认为信息扩散和关联交易是股价过度联动现象的内在因素。不同股票具有相同大股东这一关系而使股票间互相联结而形成股票网络,大股东联结的股票网络一方面反映了大股东信息扩散路径,另一方面也反映了大股东在组合投资中进行股票配置的关联交易关系。这种由于投资者关系而形成特定信息扩散路径和关联交易将导致股票间的过度联动。本文以2008年度、2009年度以及2010年度沪深300样本股相关数据进行实证分析。采用股票相关系数矩阵表达股票联动,将联动关系研究从个体间直接关系拓展到群体网络路径关系层面。以股票相关系数矩阵为因变量,以股东联结关系矩阵、股票行业关系矩阵、规模差异矩阵、价格差异矩阵、市盈率相似性矩阵等为解释变量,构建了一个股票群体网络联动关系矩阵的解释模型,着重考察股东联结关系矩阵对股票联动的影响。为了进一步考察网络位置对股票联动影响力的影响,以某股票与其它股票的平均相关系数为因变量,股票网络中心度、公司规模、价格、每股收益以及市盈率为解释变量的线性模型,讨论不同网络位置个股的市场影响力。另一方面,本文对股东联结股票网络进行了中心性和核心-边缘等网络特征分析。实证结果表明：股东联结网络对股价联动关系有显著正向影响,股票间的股东联结越强,则股票间的相关系数越大,股东联结网络反映了股票间关联交易联结和信息扩散路径而影响股价联动关系,同时行业、公司规模和市盈率特征也是股价联动关系的影响因素,同行业、公司规模和市盈率特征越相似的公司,其股价联动越紧密。进一步,越处于股东联结网络中心位置、价格越低和每股收益越高的股票,其股价走势对其他股票股价走势的影响更大。此外,通过中心性等网络特征分析,发现我国股票市场中,沪市股票在沪深300股指的影响力大于深市股票,在其中发挥着更加重要的作用。实证结果不仅验证了股东联结网络的股价联动效应,也证实了个股由于网络位置差异将导致其在股价联动中具有不同的主导地位。实证结果为股价联动的信息扩散路径和关联交易观点提供了支持,并意味着投资者组合配置内生性影响股票市场的联动关系结构。
[Abstract]:Stock price correlation plays an important role in risk management and investment decision. That is to say, the correlation of stock price movements is derived from the fundamental factors that affect the cash flow of assets and the discount rate of assets. Previous studies have shown that the degree of correlation between securities far exceeds the influence of economic fundamentals. This kind of excessive linkage has an important impact on financial theory and practice and has attracted wide attention. Existing studies believe that market segmentation of institutional investors is an important factor leading to excessive linkage. However, there is still a lack of understanding of the micro mechanism of how market segmentation leads to excessive linkage. This paper discusses the influence of investors on stock price linkage from the perspective of information diffusion and related party transactions. It is considered that information diffusion and related party transaction are the intrinsic factors of excessive linkage of stock price. Different stocks have the same relationship with large shareholders, which makes the stock connect with each other and form a stock network. On the one hand, the stock network connected by large shareholders reflects the information diffusion path of large shareholders. On the other hand, it also reflects the affiliated transaction relationship in which large shareholders allocate stocks in portfolio investment, which forms a specific information diffusion path due to investor relationship and leads to excessive linkage between stocks. Based on the data of Shanghai and Shenzhen 300 sample stocks in 2008 and 2010, this paper uses the stock correlation coefficient matrix to express the stock linkage. The research of linkage relation is extended from the direct relationship between individuals to the level of group network path relationship. Taking the stock correlation coefficient matrix as dependent variable, the stockholder connection matrix, the stock industry relation matrix, the scale difference matrix, the price difference matrix. The similarity matrix of price-earnings ratio is an explanatory variable, and an explanatory model of linkage relation matrix of stock group network is constructed. In order to further investigate the influence of network location on stock linkage, the average correlation coefficient between a stock and other stocks is taken as a dependent variable, and the center degree of stock network is taken as a dependent variable. The linear model of company size, price, earnings per share and price-earnings ratio as explanatory variables discusses the market influence of individual stocks in different network locations, on the other hand, This paper analyzes the characteristics of stockholder-connected stock network, such as centrality and core-edge. The empirical results show that: the stockholder linkage network has a significant positive impact on the stock price linkage, the stronger the shareholder connection between stocks, the greater the correlation coefficient between stocks. The connection network of shareholders reflects the linkage between stocks and information diffusion, which affects the linkage of stock price. At the same time, the characteristics of industry, company size and price-earnings ratio are also the influencing factors of the linkage of stock price. The more similar the size of the company to the price-earnings ratio, the more closely the stock price is linked. Further, the more in the center of the shareholder network, the lower the price and the higher the earnings per share. In addition, through the analysis of network characteristics such as centrality, it is found that the Shanghai stock market has more influence on the Shanghai and Shenzhen 300 stock indexes than on the Shenzhen stock market. In which to play a more important role. The empirical results not only verify the stock price linkage effect, It is also proved that individual stock has different dominant position in stock price linkage due to the difference of network location. The empirical results support the information diffusion path of stock price linkage and the viewpoint of related party transaction. It also means that the allocation of investor portfolio endogenously influences the linkage structure of the stock market.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】