Research On Investor Sentiment And Stock Market Performance Based On Microblog

With the popularization of the internet,social media has been playing an important role in our lives,which changes our roles from receivers to creators during information transmission.This motivates users of internet and generates enormous amounts of original data daily.Particularly,most of these data are people's emotion to natural and social environment.Meanwhile,most of the investors in Chinese stock market are inexperienced and irrational,which results in that the action of investors is more sensitive to emotion and affects the stock market.Thus,studying the relationship between the stock market and the investors' emotion which is acquired from the popular social media platform,is able to play an important role in risk prevention and decision making.In this paper,we adopt the microblog data from Sina and use the Tengxun Wenzhi sentiment analysis system to analyze the emotion probability,such that the emotion index can be built.We study the relationship between the emotion index and the stock market indicator by using the sorting technique.Firstly,we discuss the relationship between the stock market indicator(closing price,trading volume and volatility)and the investors' emotion in Shanghai and Shenzhen stock market respectively.We find the stock market indicator is correlated strongly to the emotion index.Afterwards,we analyze the relationship based on the selected sample of SSE 180 and find:(1)The relationship(between investors' emotion and closing price)is significant in value stock,while the relationships(between investors' emotion and trading volume,volatility)are significant in growth stock;(2)Although both of the relationships are positively related in stocks of large-scale and small-scale company,the relationship of small-scale company is more significant than large-scale company.(3)Additionally,the relationships are positively related in most of the industries,each industry suffers different levels of influence on investors' emotion.The relationships are more significant in bear market than those in bull market,even if the relationships are positively related in both bear and bull market.Secondly,we select sample of SSE 180 to discuss the impact of different investors on the relationship between the investors' emotion and stock market,and find:(1)The relationships(between the emotion of investors of different gender andeach stock market indicator)are positively related;(2)The relationships are more sensitive to female investors,and the relationships corresponding to women are more significant than men;(3)The more companies registered in a city,the more investors care about the SSE 180,and the relationships are more significant in thoese city.At last,we conduct the causality test for certifying the relationships(between investors' emotion and each stock market indicator).By conducting the Unidirectional causality test and the Ganger test based on daily and minutely sample respectively,we find the investors' emotion affects the stock market.Additionally,we find that there exits the Momentum effect in Chinese stock market,after a test of thirty lags.Our research is beneficial to investors' decision making and improving the regulatory.