Research On The Correlation Between Investment Sentiment And Stock Price Fluctuation

Under the background of financial globalization,China’s financial market has witnessed great development.The particularity of China’s securities market dominated by retail investors has led to extreme financial phenomena,such as " herding effect".which has brought great challenges to the security of the financial system,and also increased the difficulty of supervision of the financial market.Based on the indexes of stock websites,this paper selects the Psychological Line Index,Relative Strength Index,Trading Volume and Bull And Bear Index to construct the index sentiment,This process is carried out by principal component analysis;Then,the text mining technology is used to introduce online public opinion and establish text sentiment based on the text information of the forum of Oriental Fortune.Finally,according to the proportion of Chinese investors’ structure,these two sentiment indexes are combined,so that the comprehensive index can more comprehensively and accurately represent the overall investor sentiment of Chinese stock market.Taking February 27,2018 to March 16,2020 as the study interval,The model is used to divide the data into three periods: bear market,shock period and bull market.In the study of the correlation between the stock yield and sentiment,the large-cap index was studied first,and then I use the cross-correlation analysis method to study the correlation based on the Shanghai Composite Index.The research analysis shows that there are cross-correlation and multifractal characteristics between the yield of the Shanghai Composite Index and the sentiment index in the three periods,and in the bear market period,A change in one variable will have a more significant impact on another variable.In order to further study the correlation between sentiment and stock price,this paper has studied the sub-sectors,and selected the representative stocks of 47 industry sectors such as insurance,banking,real estate,etc.First of all,a single-layer network of sentiment and stock price is established to obtain the individual stocks whose sentiment(stock price)fluctuations have a greater impact on other individual stocks in different periods,indicating that there is a correlation between the sentiment(stock price)of individual stocks.Further establish the sentiment-stock price double layer network,found the most influential plate,and then analyze the index data of the double layer network.The result shows that the connectivity and connection of the double layer sentiment-stock price correlation network in the bull market period is stronger,which means that a certain impact spreads faster on the network in the bull market period.By studying the relationship between investor sentiment and the operation of the stock market,we can understand the volatility of the stock market,promote the healthy and orderly development of the financial market,and provide a reasonable theoretical basis for regulators.