A Model For Prediction Of Crude Oil Price Using Adaptive Neuro-Fuzzy Inference System

Crude oil, as a central commercial commodity, is tied to major economic activities in all nations. Also its price fluctuation plays a significant role in global political and economic situation, as a change in the price invariably affects the cost of other goods and services. Therefore reliable forecasts of the price of oil are interest for a wide range of applications. Several approaches have been proposed over the past few years. These may divide into three categories; artificial intelligent, statistical techniques and hybrid methods. The neuro-fuzzy is one of the sub divisions of hybrid method, containing neural network and fuzzy techniques. In this paper a novel forecasting model based on adaptive neuro-fuzzy inference (ANFIS) is proposed to predict monthly crude oil price. To be able to train and test the model, the statistical data from 1990 to 2010, was obtained from West Texas Intermediate oil prices. The results are compared with a multi criteria regression analysis, it is demonstrated that the neuro-fuzzy lead an accurate and more promising forecast.