Abstract: In this study we present a best model of forecasting the Gulf Cooperation Council (GCC) stock market by combining FastICA, TnA and BP algorithms (FastICA-BP) model. In this model, FastICA is firstly used to analyse the raw dataset to get components which are independent of each other. Secondly, TnA approach applied to identify and remove the IC representing the noise. And finally, BP algorithm used to predict the stock market using the filtered components. To evaluate the performance of the proposed model, Al Rajihy Islamic bank used as illustrative example in this study. The experimental results show that the proposed model outperforms the BP model using original data set (Model I) and BP model using non-filtered dataset (Model II).
Yasen Rajihy and Fadheela Sabri Abu-Almash, 2016. Combining Fastica with Back Propagation Algorithms for the Forecasting of Gulf Cooperation Council Stock Market. Research Journal of Applied Sciences, 11: 1069-1075.