Forecasting price trends and fluctuation is always a challenge for traders and investors in oil markets. Price forecasting is therefore considered as essential and practical topic. Since the different factors like, crude oil production, political events, etc, affect oil price process, forecasting oil prices have great uncertainties. On the other hand, there are various theoretical methods for modeling the price of oil. Given that forecasting is not correct usually and it has an error, so accuracy of forecasting is considered among the most important factors in the selection of forecasting procedure. Due to lack of technique agreed by researchers to forecast the oil price and owing to the difficulty in accurate identification of linear and nonlinear patterns in economic and financial time series including crude oil price, the study sought to evaluate and compare the linear and nonlinear methods using autoregressive integrated moving average (ARIMA) and nonlinear autoregressive (NAR) neural network. The results indicate the superiority of ARIMA model compared to nonlinear autoregressive (NAR) neural network in forecasting monthly values of OPEC crude oil price.