Short-Term Stock Forecasting Based on Phase Space Reconstruction and Cluster Analysis

In view of the very significant nonlinear correlation in the stock market, the stock price forecasting of stock time series is based on chaos theory and cluster analysis. The embedded dimension and time delay are preliminarily determined by the CC algorithm of reconstructing phase space. The embedding dimension and time delay are corrected by GP algorithm. The similar phase points are classified into one class by K-means algorithm and placed in the prediction model respectively. The results of the Shanghai Stock Exchange Index and individual stocks show that compared with the traditional forecasting model, the forecasting error of the model is small and the prediction accuracy is significantly improved. The validity and practicability of the model in the prediction field is proved.