Computational Neural Network for Global Stock Indexes Prediction

In this paper, computational data mining methodology was used to predict four major stock market indexes. Two learning algorithms including Linear Regression and Neural Network Standard Back Propagation (SBP) were tested and compared. The models were trained from two years of historical data from January 2006 to December 2007 in order to predict the major stock prices indexes in the United States, Europe, China and Hong Kong. The performance of these prediction models was evaluated using two widely used statistical metrics. The comparison showed that using Neural Network Standard Back Propagation algorithm resulted in better prediction accuracy then using linear regression algorithm. In addition, traditional knowledge shows that a longer training period with more training data could help to build a more accurate prediction model. However, as the stock market in China has been highly fluctuating in the past two years, this paper shows that data collected from a closer and shorter period could help to reduce the prediction error for such highly speculated fast changing environment. Index Terms Neural network, Computational Data Mining, Stock Market Forecast