Abstract
The purpose of this research is to propose an alternative framework that can meet the needs of the real-world practical application of crude oil price prediction. This study presents an alternative model based on a neural network and genetic algorithm (neuro-genetic) for the prediction of crude oil price while considering the impact of uncertainties. The model was able to learn patterns from volatile crude oil price datasets that were distorted by the Gulf War, Asian financial crises, Iraq War, Venezuelan unrest and global financial crises. The crude oil price predicted by the neuro-genetic model and the actual price were found to be statistically equal. The results obtained by the neuro-genetic model are significantly better than those of the comparison methods in terms of both accuracy and CPU processing time. The model has the potential for realistic, practical application in the real world.