Application of Genetic Algorithm to Optimization Neural Network Architecture in Predicting Oil Prices (GADDN)

The correct forecasting of oil prices plays an important role in conducting the monetary policy of different nations. Its importance is obvious in the countries exporting and importing oil. This paper uses the genetic algorithm to optimize the neural network model and structure. During the optimization process, the weights, biases and structure of the neural network are calculated. The aim of this work is reducing the complexities of using artificial neural network in forecasting. The performance of the proposed method (GADNN) is investigated by applying it WTI oil price time series prediction problems from 2012 to the end of 2015. The results show better performance and accuracy when we compare it with other neural networks model in prediction oil price.