The Influence Of Investors' Social Network Structure On The Information Efficiency Of Stock Market

After more than 30 years of rapid development,China's securities market has leapt from the "experimental field" of economic reform to the second largest market in the world.It has become an important part of China's financial system which plays a great role in promoting macroeconomic stability,supporting the development of the real economy,and improving the efficiency of capital financing.The smooth operation of the securities market is closely related to the efficient,accurate,and timely information acquisition,information transmission and information interpretation capabilities of market participants.Based on classic financial theory,the criterion for effective market is that all information is reflected in stock prices in a timely and accurate manner.The resulting equilibrium price can be expressed as the sum of the market value feedback of the information held by investors and the game value of various investors' trading behaviors.However,in the real situation of highly dense information and complex network relationships,the asset price determination mechanism shows certain limitations under the conditions of rational independence of investors and homogeneous information resources.In fact,the utility functions of investors with social attributes are state-dependent,and trading behavior is not only constrained by investors' own information resource endowment and cognitive ability,but also affected by the decision-making judgments of other economic individuals in the same external circle.Especially in the context of the rapid development of Internet information interactive media,netizens and stockholder groups are highly integrated,and the spread of financial information is accelerating,which has a profound impact on investors' learning cognition,trading mode,risk control and other aspects.Relying on the theoretical basis of behavioral finance and social finance,based on the embedded theory of strong and weak social network relations proposed by scholars such as Granovite and Bian Yanjie,this dissertation uses social network analysis methods to construct a strong relation between institutional investors' heavy holdings,and a weak relation network based on the joint questions of individual investors on the network information interaction platform.Then,taking the role of investor social network information transmission and the relationship embedding characteristics of the network structure as the starting point,this dissertation discusses the influence mechanism of different types of investor social network embedding methods,mining network node relative position,node relationship strength,and network interaction on information efficiency.It is of great significance to identify the network factors that determine asset prices,expand the interdisciplinary application of social network theory,and promote the stable and healthy development of the securities market.The main body of this dissertation includes the following three aspects:The first is theoretical model and simulation.Firstly,relying on the theoretical model of investors' rational expectations,explore the functional relationship between investor's social network centrality and information efficiency under the framework of the dual investor structure;Secondly,the extended two-dimensional cellular automata WYFW model is used to simulate the artificial stock market,which intuitively shows the heterogeneous market feedback results of investors' social network type,random information dissemination and irrational investors' conformity psychology,so as to lay a theoretical foundation for the empirical research.The second is the impact of institutional investors' social network structure characteristics on the information efficiency of the stock market.Start with the supervision effect(voting by hand)and information effect(voting by foot)of institutional investors influencing asset pricing efficiency,refine the role path of institutional investors' network centrality on the information efficiency of the stock market,and introduce corporate internal governance adjustments variables,construct a moderated intermediary effect model,and clarify the role of institutional investors as external shareholders in optimizing the company's external information environment.At the same time,expand the research on the trading behavior of institutional investors' position changes,construct a symbolic network with dynamic characteristics of holding similarity,analyze the relationship between the weighted aggregation coefficient of network residuals and the fluctuation of stock price characteristics,and determine whether trading behavior is the internal cause of stock price contagion effect.The third is to test the impact of individual investors' social network structure characteristics on the information efficiency of the stock market.Use web crawler technology to obtain the "User-Secretary of the Board of Directors" question and answer texts of SZSE's "Hudongyi" and SSE's "e-Hudong" investor information interactive platform,and then build a personal investor information network based on the link to the stock data of users' common questions,and calculate the comprehensive central indicators of important nodes in the network by using the TOPSIS multi-attribute decision-making comprehensive evaluation method.This dissertation analyze the effect of individual investor network centrality on the economic dimension of information efficiency(private information:stock price information content)and accounting dimension(public information:accuracy of earnings forecast).On this basis,introduce the centrality of institutional investor network and institutional investor's shareholding ratio as two regulatory variables,then analyze the heterogeneous influence of two types of network centrality interaction on market information efficiency from the perspective of external shareholders' consistent actors.Through the research above,the main conclusions are as follows:The first is the relationship between institutional investors' social network structure and information efficiency:(1)Institutional investor network centrality has a positive effect on the information efficiency of the stock market.The relative positional advantages of network nodes can enhance the information content of stock prices and increase the speed of stock price integration into market information.Based on the classification of investment style and investment period,prudent and long-term institutional investors can help improve the information content.There is no significant difference between the two classification methods on the speed of information transmission,indicating that various types of institutional investors can play an active role in accelerating information transmission;(2)Institutional investors located in the center of the network mainly improve information efficiency through two intermediary paths:improving the level of information sharing and improving the company's entrusted agency problem.Based on the perspective of information effects,institutional investors in core locations tend to share information rather than information block or competition,in order to form a collusion to correct irrational price deviations and realize the discovery of the company's intrinsic value;from the perspective of supervisory effect,institutional investors in core positions,under the incentive of performance ranking and status reputation,are likely to form checks and balances with the company's internal shareholders,exerting the function of external shareholder supervision,restrict the company to cater to earnings manipulation,so as to improve the quality of accounting information and information transparency;(3)Good internal governance of the company plays a regulatory role,reducing the cost of obtaining information from external investors,improving information transparency,and narrowing the information gap between the "center-edge" nodes of the network;(4)The residual weighted clustering coefficient of the similarity network of position change is positively correlated with information efficiency,indicating that the trading strategy of institutional investors jointly opening or closing positions is derived from the feedback of corporate value information rather than noise trading,which proves the positive effect of institutional investor network structure on stabilizing stock price and improving pricing efficiency.The second is the relationship between the social network structure of individual investors and information efficiency:(1)From the economic perspective:the centrality of individual investor information network and the information content of stock prices are in an "inverted U-shaped" relationship.Based on the two-level communication theory,when the centrality of the network is located before the U-shape's turning point,the core nodes play the role of "opinion leaders",reducing the cost of obtaining information from nodes in the same circle,attracting the attention of external circles,and reaching a consensus,forming a combined supervision force,which will force the company to improve the quality of information disclosure;when the network centrality exceeds the U-shape's inflection point,the self-reinforcing psychological bias of the core nodes induces the coloop nodes to excessively pursue group identity,and the limited attention makes their ability to process overloaded information diminishes marginally,leading to selective bias and noise trading,and thus reducing the efficiency of market information.At the same time,based on the classification of interactive information content,if the question text contains the company's medium-term macro information and long-term development strategy keywords,investors will focus on the company's long-term value discovery,high-quality information feedbacks will strengthen the network centrality and the non-synchronization of stock prices.Conversely,if investors pay attention to company emergencies and notify short-term information such as announcements,the above-mentioned non-linear relationship will be weakened;(2)From the accounting perspective:the core position node plays a positive role in the correction of unanticipated earnings and effectively improves the accuracy of earnings forecast.The third is the long-term and short-term impact of the interaction of the two types of investor social networks on information efficiency:the centrality and shareholding ratio of institutional investors' networks have a supplementary effect on the information transmission function of individual investor networks.Therefore,under the same disclosure environment of listed companies,the "inverted U-shaped" relationship between the network centrality of individual investors and the non-synchronicity of stock price fluctuations,as well as its correction effect on unexpected earnings,increases with the decrease of the centrality(shareholding ratio)of institutional investors.The innovations and main contributions of this dissertation are reflected in the following three aspects:Firstly,in terms of social network analysis methods,given the fact of the high proportion of small and medium investors in the Chinese securities market and the high coupling of the identities of stockholders and netizens,we established the weak connection relationship of individual investor network by using the data of users on the SZSE's"Hudongyi" and SSE's "e-Hudong" investor information interactive platforms creatively,which provides an important supplement to the traditional strong relationship social network analysis and provides evidence support for the application of social network analysis method to the research on problems in the financial field.Secondly,in terms of the research framework of information efficiency,based on the analysis of various investor game behaviors,we indicatively examine the interaction of the two types of investor networks on the governance and information disclosure level of listed companies from the perspective of external small and medium shareholders' consistency.Which will greatly enrich the research content of investor heterogeneity,as well as provide a more comprehensive analysis of the research on market information efficiency.Thirdly,In terms of the policy effects of the research conclusions,China's securities market is in a critical period of transition from high-speed to high-quality development,and the problem of financial anomalies caused by investors' irrational behavior needs to be resolved urgently.Through the interdisciplinary and multi-perspective analysis of the micro-mechanism that affects the efficiency of the securities market,this dissertation excavates the network driving forces of asset price changes.The research conclusions have important theoretical value and practical significance for improving the pricing efficiency of China's securities market and the level of information disclosure of listed companies.