The Influence And Mechanism Of Social Media Information On The Stock Market

The information of social media not only impacts the information environment of the stock market,but also provides a new perspective for scholars to study the market efficiency.Its relationship with the stock market has become one of the hot issues of current researches.To this end,this research focuses on the social media platform under the financial background,analyzes the mechanism of its published information and the stock market,and combines the micro-investor behavior analysis with macro-platform role's research.In terms of micro perspective,this paper studies how investors integrate emotional information and other textual information published by social media platforms simultaneously to make investment decisions;and how investor interactions affect the liquidity of stock markets.In terms of macro perspective,this research integrates social media information with other information that most investors that easily accessing together and analyzes the intermediary role of social media information in stock market volatility.After systematic review of the literature on the relationship between social media and stock market at home and abroad,it is found that current research mainly focuses on the "single" relationship between certain characteristics of social media information and stock market,such as the relationship between page views and stock market,or the relation between emotion and stock market,etc.There are rare literatures illustrate the interaction of different social media information characteristics in stock volatility.Meanwhile,most of existing studies ignore the role as “bridge” of social media information in the process of participating in other stock volatility caused by information.Therefore,this paper starts from above two dimensions,and makes a series of studies on social media's influence on the stock market and its mechanism of action.The questions of this paper include:(1)Does social media emotion have an impact on individual stock returns? Can the central path feature variable and the edge path feature variable extracted from passage adjust the relationship between social media emotion and individual stock returns?(2)Does social media user interaction have an impact on stock liquidity? For companies with serious information asymmetry problem,will this effect be more obvious? Can the textual features extracted from social media texts adjust the relationship between user interaction and stock liquidity?(3)Does news or analyst emotion affect individual stock returns in the capital market? If news or analyst emotion has an impact on individual stock returns,then whether and how do above two emotions “shape” the social media sentiment,and affect overall stock market? The main research contents,conclusions and contributions of this research are as follows:Study 1.The impact of social media information on stock market--a moderator effect based on text features.This part mainly discusses two issues,the first is the impact of social media emotion on individual stock returns.The second is the central path feature variables extracted from social media texts and the moderating effect of edge path feature variables on social media emotions and individual stock returns.This part refers to the ELM theory,under the background of GEM market in China,it extracts the text features of central path and marginal path based on the sample of more than 1 million post information provided by the eastmoney.com(300 stocks between March 2018 and April 2019),and constructs a fixed effect model of panel.The empirical test examines the extracted center features of social media text and the regulation effect hypothesis of the edge path features on the relationship between emotion and individual stock returns.The research results show that the textual emotion of social media has a significant positive impact on individual stock returns;In addition to the professional features,all other central path feature variables(fog index and evidence support)have a significant negative adjustment effect on the relationship between emotion and individual stock returns;All edge path feature variables(language strength,text length)extracted from social media information have a significant positive adjustment effect on the relationship between emotion and individual stock returns.Main contributions: This part of the research focuses on the central features of social media texts and the role of marginal feature variables in the relationship between social media emotion and individual stock returns,which has important theoretical and practical value.At the same time,for the first time,from the perspective of investors,it illustrates the effect of social media emotion and other textual information on the securities market and the mechanism behind it,and extends the connotation and significance of the social media emotion's influence on the stock market.Study 2.The impact of social media information on stock liquidity--a moderator effect based on text features.Under the background of GEM market in China,this paper mainly investigates the relationship between social media user interaction and stock liquidity based on the research object of more than 4 million posts provided by the eastmoney.com(345 stocks between October 2012 and September 2015),and analyzes the difference between the user interaction influence on stock liquidity affecting the differences between high and low information asymmetry companies.In addition,based on the feature extraction of social media texts from the previous question,the paper analyzes the role of text features in adjusting the social media user interaction and stock liquidity.The results show: Social media user interaction has a significant positive effect on stock liquidity;The higher the degree of information asymmetry of listed companies,the stronger the role of social media user interaction in promoting stock liquidity;The professionalism,evidence support,and text length features of social media texts have a significant positive adjustment effect on the relationship between user interaction and stock liquidity;The fog index and text strength features of social media texts have a significant negative adjustment effect on the relationship between user interaction and stock liquidity.Main contributions: This part of the research combines the current background of social media's rapid development,and uses social media data to analyze the internal mechanism of stock market flow,which broadens the boundaries of original research to certain degree.Meanwhile,this research focuses on the role of different text features in social media user interaction and stock liquidity for the first time,and discusses different effects of different text features on reducing information bias and asymmetry between the investors,which are conducive to understanding the mechanism of investors' willingness to trade,thus better explaining the internal mechanism of stock market circulation.Study 3.How news sentiment and analyst sentiment affect stock returns--a moderator effect based on social media sentiment.This part collects the news reports(131,866 news data)and analyst rating data(68,640 analysts rating data)on 345 stocks of the GEM from October 2012 to September 2015 respectively,and extracts their respective emotions.By reviewing the literature and focusing on the "Agenda Setting Theory" and "Theory Credibility Theory",the test examines the mediating effects of social media emotion between news emotion and analyst emotion and stock returns.The results show that the public emotion released by social media plays a partial intermediary role in the process of news emotion affecting individual stock returns.This indicates that in the process,it affects the public emotion at least partly through the “algorithm setting function”,which means social media emotion has intermediary effect;the public emotion released by social media plays a partial intermediary role in the process of analysts' emotions affecting individual stock returns.This means that in the process,it affects the way and effect of public emotion at least partly through the “resource effect reliability effect”,which means social media emotion has some mediating effect.Main contributions: This part uses social media text information-emotion as an entry point,introduces a bridge of public emotion,analyzes the news emotions or analyst emotions simultaneously with social media emotion,and investigate the role of social media emotion in the stock market.It not only verifies the research on the market influence of stocks and analysts' reports,but also investigates the mediating role of social media emotion in the process of news or analysts' influence on individual stocks,which enriches relevant theories of the relationship between social media information and the stock market,expands its theoretical mechanism of action mechanism and conduction path with the stock market,providing new ideas and methods for research in this field.