Changes in the stock price index can effectively reflect the price level of various stocks in the stock market and help investors understand the circumstances of the stock changes.At present,the role of China's stock market in the national economy is becoming more and more important.Therefore,the study of the rule of stock index changes is essential for both the market managers and the majority of investors.To predict the change rule of the stock index,this paper establishes the regression model.First of all,it should start from the selection of experimental data.Different from the previous experiments,we use a variety of experimental data,including the market index and stock historical transaction data.This is because that the stock index is received by the stock price weighting and through understanding the stock transaction data,people can more effectively obtain the rule of stock index changes.Secondly,we propose some new eigenvector solution methods.The eigenvector is a very important factor in the accuracy prediction of the model.We obtain from different aspects,such as the transaction distribution,the transaction contrast,general comparison of upward and downward trend the function of rising and falling,the kinetic energy trend etc.Finally,according to the characteristics of sample,we choose the support vector regression as our prediction model,and use the genetic algorithm to optimize the parameters.Experiments show that the model based on this paper can effectively predict the change of stock index.Also this model plays a significant role and it can help the majority of investors better understand the changes in the stock market. |