International Oil Price Time Series Prediction Using GMDH Neural Network and its Performance Comparison with MLP Neural Network and ARIMA Method

Authors

  • Alireza Haji , Mahdi Ghazanfari, Erfan Rafie Kia

Abstract

Predicting oil prices, especially in exporting countries, will help governments in the policy-making process by obtaining a reliable estimate of oil revenues. The existence of a complex mechanism governing the process of oil price formation has reduced the efficiency of linear models in forecasting and led researchers to use nonlinear intelligent systems to predict oil prices. In this study, after a detailed study of the structure of an artificial neural network, two models of neural network GMDH and MLP and ARIMA method has been used to predict oil prices. There are important factors in the prediction process with neural networks, and if all these factors are selected correctly; One can expect the neural network to have a good prediction. In this research, in addition to oil prices, the variables of global oil production and consumption and commercial oil reserves of OECD countries have been tried to be used as input variables for forecasting; And variables that have higher predictive power should be kept in the model. Neural network error and Arima method are compared with the mean squared error (MSE) criterion. Finally, the results indicate that the MLP neural network model provides a more accurate prediction than the other two models And has higher accuracy.

Published

2021-07-01

Issue

Section

Articles