Neural network technology for stock market index prediction

Two neural network models, the radial basis function (RBF) and backpropagation, applied to stock market index predictions are compared. Actual data of the Wall Street Journal's Dow Jones Industrial Index has been used for a benchmark in the experiments. A notable success has been achieved with the proposed models producing over 80% prediction accuracies observed based on the monthly Dow Jones Industrial Index predictions. These models have also captured both moderate and heavy index fluctuations. The experiments conducted in this study demonstrated that the RBF neural network is preferred over the multilayer perceptron network and is a promising candidate for stock market index predictions.<<ETX>>