Research On The Influence Of Investor Sentiment On The Return Rate Of China's Stock Market

The development of various online social platforms such as Weibo,blogs,WeChat,and forums means that we have entered a new era of the Internet.The online social platforms play an increasingly important role in people's lives.It changes the way people transmit and share information.People can express their opinions on social issues through the online social platforms anytime and anywhere.With the advancement of text mining technology and the arrival of the era of big data,web crawlers provide powerful technical support for accessing massive amounts of network information.Especially for China's current capital market,there are many irrational investors in the stock market,and their investment behaviors are affected by subjective emotional factors,making the Chinese stock market full of irrational emotional trading.Behavioral finance theory shows that investor sentiment has a significant impact on the development of capital markets,but it is controversial about how to measure investor sentiment.Traditional investor sentiment indicators cannot fully reflect the psychological state of investment entities,but,based on social networks,the investor sentiment provides a new channel for measuring investor sentiment.For example,financial media and economic experts publish stock market information in real time on the online platform.Investors timely obtain the information and express their views on the market and individual stocks.These data represent the judgement of the whole market towards the current and future periods,making a great influence on the stock market trend.Therefore,using the text data of the online social platform to construct investor sentiment indicators and explore the relationship between investor sentiment and the stock market,having important research value for China's stock market.As an important application in the social networking platform,Sina Weibo's universality of users,rapid spread of information,and self-production of content has made it an important information communication medium.Since more and more financial media registering accounts on Weibo and publishing stock market information through Weibo in real time,this article uses Sina Weibo as a data source,crawl information related to the stock market by keywords,and obtain a total of 4.2 million microblogs from 2015 to 2018.With the textual sentiment analysis technology,the crawled microblog text data can be quantified into three categories,including the positive emotions,neutral emotions and negative emotions,which helps to construct the microblog investor sentiment indicators.The total market value and the after-one-day return rate were selected as the control variables,and the quantile regression model was used to empirically analyze the yield of the Shanghai Composite Index.The empirical results confirm that there is a significant impact between the investor sentiment indicators based on Sina Weibo text analysis and the stock market.The empirical results also show that there is a significant correlation between microblogging sentiment and stock market returns in some specific quantile intervals,which means that Weibo investor sentiment has certain predictive ability on the conditional quantile of stock market returns.The emotional changes predict the tail behavior characteristics of stock market returns and provide decision-making reference for financial risk prevention.