基于持股关系的股票市场关联网络研究
发布时间：2018-08-19 06:03
【摘要】：随着股票市场的发展,上市公司数量越来越多,上市公司及其股东间的投资关系越来越复杂,在股票市场各种参与者的共同作用下,股市从整体上表现出了与单个参与者完全不同的特征。对股票市场参与主体构成的网络进行研究可以从整体上把握股票市场的发展变化特征,从而有助于及时发现股市中存在的问题,为股市监管及投资者带来启示。因此,有必要对股票市场的复杂特性进行整体研究。在国内外相关文献调研的基础上,本文首先对与股票市场关联网络形成有关的理论及复杂网络的理论与方法进行了介绍。其次明确了本文所研究网络的构建方式——以上市公司及其十大股东为节点,节点间的实际投资关系为边,投资额大小为边的权重构建了一个股权关联网络。再次运用CCER数据库收集了2001至2010年十年沪深两市所有上市公司及其十大股东的持股数据,并利用Pajek、Cfinder等分析软件对各年份以及2010年各行业的股权关联网络进行了分析,最终得出以下有价值的结论：沪深两市股权关联网络具有明显的无标度和小世界特征,大部分资金集中在少数的吸引投资能力强的上市公司手中,同时大部分股权集中在少数投资能力强的股东手中,任意两个上市公司或者股东间通过较少的投资关系就能进行联系；不同行业股权关联网络的无标度特征存在一定的差异,农林牧渔业的无标度特征最为明显；沪深两市中最大的投资者为上市公司控股集团公司或者母公司、上市公司本身、国有专业投资机构、社保基金以及外资投资,其中外资投资具有明显的行业偏好特性,大额的外资投资主要集中在采掘业、交通运输仓储业和金融保险业；网络中存在大量的社团,而且大额度投资之间的联系更为紧密,在2008年全球金融危机的情况下,更多的公司选择多角化投资,社团中节点数量突增；金融行业内部的投资关系最为紧密,在金融行业中也存在比其他行业都大的社团。最后,根据前文得出的结论,得到一些股票市场监管及投资者投资方面的启示及建议。本文的贡献之处在于：打破以往研究股票市场关联网络只是以上市公司作为节点,将股票之间的价格波动或者收益率波动的相关系数作为边权重的限制,将上市公司及其股东都包括入所构建的网络中,并以真实的投资关系反映网络中节点的联系,用定量分析的方法探析了沪深两市总体以及各个行业的网络结构特征,并得出了一些有用的启示及建议。例如：在股票市场的监管过程中可以借鉴复杂网络中“目标免疫”的方法；投资者进行投资分析的过程中可以将外资投资作为参考等。
[Abstract]:With the development of the stock market, the number of listed companies is increasing, and the investment relationship between listed companies and their shareholders is becoming more and more complicated. The stock market as a whole shows a completely different feature from that of a single participant. The research on the network of the main body involved in the stock market can grasp the characteristics of the development and change of the stock market as a whole, thus help to discover the existing problems in the stock market in time, and bring enlightenment to the stock market supervision and investors. Therefore, it is necessary to study the complex characteristics of stock market as a whole. On the basis of domestic and foreign literature, this paper first introduces the theory and method of complex network related to the formation of the related network in the stock market. Secondly, the paper makes clear the way of constructing the network-taking listed company and its ten shareholders as the nodes, the actual investment relationship between the nodes as the edge, the investment size as the edge weight to construct an equity association network. The data of all the listed companies and their ten shareholders in Shanghai and Shenzhen stock market from 2001 to 2010 are collected by using CCER database again, and the analysis software, such as Pajek Cfinder, is used to analyze the stock ownership correlation network in each year and in 2010. Finally, the following valuable conclusions are drawn: the Shanghai and Shenzhen stock market equity correlation networks have obvious scale-free and small-world characteristics, and most of the funds are concentrated in the hands of a few listed companies with strong ability to attract investment. At the same time, most of the equity is concentrated in the hands of a few shareholders with strong investment ability, any two listed companies or shareholders can be linked through less investment relationship; there are some differences in the scale-free characteristics of the equity association network in different industries. The biggest investors in the Shanghai and Shenzhen stock markets are listed companies holding groups or parent companies, listed companies themselves, state-owned professional investment institutions, social security funds and foreign investment. Among them, foreign investment has obvious characteristics of industry preference. Large amount of foreign investment is mainly concentrated in extractive industry, transportation, warehousing and finance and insurance industry. There are a large number of associations in the network, and the relationship between large amounts of investment is more close. Under the global financial crisis in 2008, more companies choose to invest in diversification, and the number of nodes in the community has increased dramatically. The investment relationship within the financial industry is the most close, and there are also larger associations in the financial industry than in other industries. Finally, according to the above conclusions, some enlightenment and suggestions on stock market supervision and investor investment are obtained. The contribution of this paper lies in: break the previous research on the stock market correlation network only take the listed company as the node, take the stock price fluctuation or the correlation coefficient of the return fluctuation as the edge weight limit, The listed companies and their shareholders are included in the network, and the real investment relationship is used to reflect the connections of the nodes in the network. The characteristics of the network structure of the Shanghai and Shenzhen stock markets as a whole and the various industries are analyzed by quantitative analysis. Some useful enlightenments and suggestions are obtained. For example, in the process of stock market supervision, we can learn from the method of "target immunity" in complex networks, and investors can take foreign investment as reference in the process of investment analysis.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F275;F832.51
本文编号：2190859
[Abstract]:With the development of the stock market, the number of listed companies is increasing, and the investment relationship between listed companies and their shareholders is becoming more and more complicated. The stock market as a whole shows a completely different feature from that of a single participant. The research on the network of the main body involved in the stock market can grasp the characteristics of the development and change of the stock market as a whole, thus help to discover the existing problems in the stock market in time, and bring enlightenment to the stock market supervision and investors. Therefore, it is necessary to study the complex characteristics of stock market as a whole. On the basis of domestic and foreign literature, this paper first introduces the theory and method of complex network related to the formation of the related network in the stock market. Secondly, the paper makes clear the way of constructing the network-taking listed company and its ten shareholders as the nodes, the actual investment relationship between the nodes as the edge, the investment size as the edge weight to construct an equity association network. The data of all the listed companies and their ten shareholders in Shanghai and Shenzhen stock market from 2001 to 2010 are collected by using CCER database again, and the analysis software, such as Pajek Cfinder, is used to analyze the stock ownership correlation network in each year and in 2010. Finally, the following valuable conclusions are drawn: the Shanghai and Shenzhen stock market equity correlation networks have obvious scale-free and small-world characteristics, and most of the funds are concentrated in the hands of a few listed companies with strong ability to attract investment. At the same time, most of the equity is concentrated in the hands of a few shareholders with strong investment ability, any two listed companies or shareholders can be linked through less investment relationship; there are some differences in the scale-free characteristics of the equity association network in different industries. The biggest investors in the Shanghai and Shenzhen stock markets are listed companies holding groups or parent companies, listed companies themselves, state-owned professional investment institutions, social security funds and foreign investment. Among them, foreign investment has obvious characteristics of industry preference. Large amount of foreign investment is mainly concentrated in extractive industry, transportation, warehousing and finance and insurance industry. There are a large number of associations in the network, and the relationship between large amounts of investment is more close. Under the global financial crisis in 2008, more companies choose to invest in diversification, and the number of nodes in the community has increased dramatically. The investment relationship within the financial industry is the most close, and there are also larger associations in the financial industry than in other industries. Finally, according to the above conclusions, some enlightenment and suggestions on stock market supervision and investor investment are obtained. The contribution of this paper lies in: break the previous research on the stock market correlation network only take the listed company as the node, take the stock price fluctuation or the correlation coefficient of the return fluctuation as the edge weight limit, The listed companies and their shareholders are included in the network, and the real investment relationship is used to reflect the connections of the nodes in the network. The characteristics of the network structure of the Shanghai and Shenzhen stock markets as a whole and the various industries are analyzed by quantitative analysis. Some useful enlightenments and suggestions are obtained. For example, in the process of stock market supervision, we can learn from the method of "target immunity" in complex networks, and investors can take foreign investment as reference in the process of investment analysis.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F275;F832.51
【共引文献】
相关期刊论文 前1条
1 吴华骢;;发行制度变迁视角下的IPO定价影响因素研究[J];中外企业家;2015年11期
相关硕士学位论文 前4条
1 赵佼;中国证券市场信托投资行为与监管研究[D];首都经济贸易大学;2011年
2 吴静;会计政策变更与股价的相关性研究[D];天津财经大学;2013年
3 朱金贺;生猪养殖者市场风险认知与预控研究[D];山东农业大学;2014年
4 郝娟;非营利组织信息披露质量的影响因素研究[D];湖南大学;2014年
，本文编号：2190859
本文链接：https://www.wllwen.com/guanlilunwen/huobilw/2190859.html
