A Stock Market Intraday Trading Strategy Based On Artificial Neutral Network Forecasting

Predicting trends in the stock market is a subject of major interest for both scholars and financial analysts.The main difficulty lies in the dynamic,complex and disorder nature of the market.In this paper,BP neural network was used to construct the forecasting model of stock price fluctuation.In addition,this paper designs a set of intraday trading system which can transform the output of artificial neural network(ANNs)into investment decision,and pointed out the best transaction timing for investors..The forecast result of ANNs model was the intraday fluctuation interval which was composed of the highest and lowest forecast price of the day.Based on this,this paper constructs an intraday trading strategy.The output of the forecast model was used as the input variable of the trading system directly.Trading system performance can also be used to test the performance of the forecasting model.This approach was not only superior to the traditional forecast error statistics,but can also help investors to quickly form investment decisions and improve their trading results.In empirical research section,both of the error indicator MSE and back-test trading performance were showed to describe the accuracy of the prediction and the result of stock trading.Then,the work extended to the study of strategy portfolio.By combining this intraday strategy with "holding strategy" or "timing strategy",investors can enhance the strategy portfolio in both profitability and risk control ability,and the conclusion can be drawn that this intraday strategy can be used as an enhancement of stock holding strategies(including alpha stock selection strategy,timing holding strategy,etc.),and it would play an important role in the portfolio.