Homogeneous Ensemble FeedForward Neural Network in CIMB Stock Price Forecasting

Abstract

In stock forecasting, the main concern is the accuracy of the stock market prediction. Various methods, techniques and algorithms have been proposed to achieve higher prediction accuracy. One of the popular techniques is Artificial Neural Network (ANN). ANN is a bio-inspired algorithm that tries to simulate the brain’s mechanism which is achieved through simulating a connected artificial neuron network. Although ANN has shown promising results in forecasting, researches have shown that combining multiple neural networks can further improve the forecasting result. The aim of this paper is to compare the performance of a single and multiple homogeneous ANN in forecasting CIMB stock market closing price. The empirical result shows that homogeneous ensemble ANN performs better than a single ANN in predicting the stock market price.