Statistical Analysis Of The Stock Data And The Prediction Algorithm

With the rapid development and expansion of the stock market,stock data ballooned.How to analyze these massive stock information and excavate its intrinsic information and value through artificial intelligence method is the concern of investors and regulators.The research work of the listed company's financial data classification and stock price forecast can not only grasp the future dynamic changes of the stock market,but also can guide the stock investors to invest reasonably and raise the profits.The main research content and innovation are as follows:1)An improved comprehensive classification evaluation method is proposed.This method replaces the original principal component scoring matrix with the single factor evaluation matrix,and replaces the original principal component weights with the variance contribution rate of the weighted principal component,and realizes the comprehensive classification and evaluation of the financial ability of the listed companies of real estate industry with the help of Fuzzy mathematics operator.2)An improved algorithm of BP neural network based on Golden Segmentation optimization algorithm is presented.Golden Segmentation optimization algorithm is used to optimize the number of hidden layer nodes,the weights of input layer and the deviation of hidden layer.The experimental results show that compared with the generalized linear regression,time-series,traditional BP neural network,support vector machine and extreme learning machine,the method can achieve higher prediction accuracy.3)An improved algorithm of support vector machines based on particle swarm optimization with active operator(APSO)is proposed.Active operator particle swarm optimization is used to optimize the kernel parameters and penalty coefficients,meanwhile the Boruta algorithm is used to extract low-dimensional features from the raw data.The experimental results show that compared with the traditional BP neural network,random forest,support vector machine and extremelearning machine,this algorithm can achieve higher classification accuracy.