Abstract:Following local wave analysis and chaos theory, a method is proposed to model and forecast stock market. First, using local wave analysis theory, Shanghai composite index (SCI) daily gain rate series are decomposed respectively into two parts: low frequency part and high frequency part and using 2-order Renyi entropy, the further analysis of chaos feature of decomposition is done. Then, by using Multi-point similarity forecast method, the chaotic forecasting models are established to forecast the low frequency part and the high frequency part respectively. Finally, the forecasting result of chaotic model is reconstructed by local wave analysis. By doing so, the forecasting of the original SCI daily gain rate series can be made. Our result demonstrates that the method is of high precision and wide potential applications.