Intelligence Mining Research On The Relationship Between Baidu Index, Financial News And Stock Price

The efficient market hypothesis holds that in an efficient market for capital,asset prices can reflect all available information.With the advent of the Internet age,the media plays an important role in the capital market.Financial news is a subcategory of financial media,which can collect,report and publish financial news about listed companies in a timely manner.Financial news can convey information in a timely manner and attract investors’ attention.Due to the different motives of financial news reports,the content and attention elements of financial news reports are different.Therefore,it is of great theoretical significance to study the relationship between Baidu index and financial news stock prices: The ability to expand the existing research is to deepen the research on the relationship between investor attention and financial news and the relationship between stock prices,and can provide reference for investors to invest.In this paper,the search volume of the Baidu index is used as the investor attention,the financial news text content and the number of news are used as the data,and the financial news of listed companies is used as the text,and then the LDA topic model is used to analyze the main content of the financial news.Gain attention by paying attention to content,use VAR model,Granger causality,and impulse response model to analyze the relationship between Baidu index and financial news and stock prices,and then further mine information.In order to make this research more accurate,this paper uses regression analysis to study the relationship and correlation between Baidu Index,financial news and stock prices.The research found that: first,the LDA topic model constructed by retrieving relevant news from listed companies can get the hot spots that the financial media pay attention to.Second,when retrieving the hot spots that investors pay attention to through the hot spots that the media pays attention to,it is found that the stock price changes correspondingly with the attention of investors.It shows that a large number of investors will follow the attention of the public to make investment strategies.Third,Granger causality test is carried out on the topic elements obtained through the LDA topic model constructed by the news text,and the investor attention data obtained through the topic elements and the stock closing price.The research found that the Baidu index and the closing price of the stock price are Granger causal to each other.Changes in the Baidu index will impact the stock price,and the rise and fall of the stock price will cause investors to pay attention to the stock price.The lag period is 1-4 periods.Fourth,through the Granger causality test of the Baidu index and the stock closing price,it is found that investors will pay more attention to the product factors and innovation factors of the company.Fifth point,through the analysis of the constructed multiple regression model,the opening price of the stock,the Baidu index,and the number of news will all affect the change of the closing price of the stock.The results show that the changes in the number of financial news and the Baidu index have a significant impact on the stock closing price.