Research On Neural Network Stock Market Trend Forecasting Algorithm Based On Window Energy Calculation

Stock market is an important means of corporate finance and investor investment.Stock market prediction research has great theoretical and practical significance for investors,enterprises and government policy making.Compared with the one-day prediction,the short-term trend prediction,especially the short-term trend of the turning point prediction whether it is for short-term investors or long-term investors have a lot of operational guidance.However,due to the lag of the technical indicators,the closing time series is often focused on the continuity of the trend,making the traditional stock price prediction model for the trend of the turning point of the lack of interpretation and poor prediction.Based on the K line combination,average combination and volume and price combination,thesis uses window energy in horizontal trend and non-horizontal trend to forecast the short-term trend of the Shanghai Composite Index.The specific research work is as follows:Horizontal trend lasts for a short time,and the uncertainty of its direction changes is huge,so it becomes hard to forecast the direction of horizontal condition's trend in Stock Market trend prediction.Based on the energy calculation of Horizontal window,A BP neural network algorithm(WE-BPNN)is presented for predicting Horizontal window trend.Firstly,the division standard for short-term trend is given,on the basis of which we comes up with definitions of horizontal window;Then,by calculating the energy of K-line combination and Moving Average combination,we merge two types of energy into window energy;At Last,leading the window energy into the direction of BP Neural network to predict window trend.Because of hysteresis of energy's influences on the trend,there is a case that energy accumulated while trend not changes,it will affect the accuracy of the trend prediction.Thus,basing on WE-BPNN neural network algorithm we lead energy regulator into BP neural network algorithm(EF-BPNN),Dynamically adjusting Weights of window energy for the trend prediction.On the Shanghai Stock's data,the experimental results show that EF-BPNN algorithm has better performance.In strong bull market and bear market,guidance of EF-BPNN algorithm is very large,but for volatile market,that is,when the persistence of trend is poor,the non-horizontal trend of sustainable prediction is even more important.Based on the phenomenon of resistance and discrepancy of volume and price in the non-horizontal trend,we put forward the concept of trend energy and discrepancy energy that is similar to the physical model of frictional resistance and kinetic energy,and introduce the calculation of trend energy and discrepancy energy.we propose trend prediction Algorithm Based on discrepancy Energy(BE-BPNN)in non-horizontal trends.Experiments show that the persistence' prediction of trend in the non-horizontal trend,both the recall rate and accuracy have a very good effect.