Oil Price Predictions Based on Nonlinear Smooth Transition Models, Using Genetic Algorithm Optimization

The price of oil plays an important role in the global economy and is an important factor influencing the government and commercial sectors. Because of increasing importance of oil in financial markets, oil price predictions have always been an important subject for the researchers in Economics, and other economic agents. This paper tries to study the behavior of crude oil prices based on Smooth Transition Autoregressive models used on monthly crude oil prices data. We show that simulation results using STAR models estimated by Genetic Algorithm method, outperforms linear time series models, such as ARIMA, for out of sample predictions based on RMSE and MAE and DA criteria.