Research On The Relationship Between Internet Financial News And Stock Price Volatility

With the advent of the Internet era,more and more people will pay attention to financial news through the Internet.There is no doubt that the majority of investors will be concerned about the trend of the stock market through financial news.And then,whether there id a link between Internet financial news and stock movements,and which categories of words has relationship with stock movements,all these are the focus of attention of investors,and but also is a focus in the study of this article.This article is based on text processing technology and regression analysis technology,as an example of Sina finance and economics website,through the web crawler,acquired all the news about margin trading stocks on the Sina finance website from to January 1,2015 to September 30,2016.By using the technology of text segmentation and quantification,the quantitative analysis of the news text is carried out,and then the quantized data are classified and classified into 6 categories.Finally,by the method of named entities to match the quantitative data of the news text with the stock price minute data,the data used in regression analysis has been obtained.Based on the independent variable of the weights of six categories words,and the dependent variable of the fluctuation range of the share price and the direction of change,multiple linear regression model was used to study the difference between the total sample and the sample of Shanghai and Shenzhen.At the same time,because of the direct effect of the stock price fluctuation is very poor,so this paper attempts to use Logistic regression to forecast the direction of stock price.Through empirical analysis,the following conclusions are obtained: On the whole,there is a certain relationship between the financial news text and the stock price fluctuation,and there are some differences between the all kinds of words and the stock price fluctuation.There is a negative relationship between the negative words and the stock price volatility,otherwise the opposite.The degree of close relationship between the proper nouns(Words of the company's operating conditions and Trade terms)and the fluctuation of the stock price is greater than that of the emotional words and the stock price volatility.Under the total samples and samples in the Shanghai market,6 categories of words will be a slight change with the passage of time,the relationship with stock price volatility of the close degree.Generally speaking,as time goes on,the degree of close relationship between stock price volatility and business status terms tends to decline,and for the trading terms,their close relationship degree is on the rise,as for the emotional words,the relationship between them has not changed much.Using Logistic regression method to financial news in the 6 categories of terms as the independent variable to predict stock prices rise and fall of the effect,analyzing the prediction effect and the reasons,providing some reference for investors and related researchers.The forecast accuracy of the changes of stock prices in Shanghai is higher than that in Shenzhen.The best prediction time is about 10 minutes,for the news text to the stock prices of financing and Securities Lending.