Brent crude oil Price Forecast with Hybrid Model of Nonlinear Grey Model and Linear Arima Waste Correction

The characteristics of crude oil and the factors affecting the price of this energy carrier have caused its price forecast to always be considered by researchers, oil market activists, governments and policy makers. Since the price of crude oil is affected by many factors, therefore, continuous studies should be done in this way so that the estimates made over time, the results are more accurate and more reliable. In this paper, a combination of nonlinear grey model and Arima is used to forecast the price of crude oil and a combined nonlinear grey-Arima model is proposed. Brent crude oil price data for seasonal, monthly and weekly periods have been used to investigate this technique. In the quarterly forecast of data for the first quarter of 2015 to the fourth quarter of 2021, in the monthly forecast of data for January 2020 to December 2021 and in the weekly forecast of data for the second quarter of March 2020 until the first week of December 2021 has been used. The results showed that the mean absolute value of the error percentage and the square root of the mean square error in the hybrid model are always lower than the single theory-based model or the nonlinear grey single theory. Also, the hybrid model has a higher ability to explain and cover price fluctuations in different time periods and is more reliable than the single model. Therefore, a hybrid model can be used instead of single and single theory-based models for more accurate forecasting. JEL Classification: Q47, C53, C01 Keywords: Crude Oil Price, Crude Oil Price Forecast, NGM Nonlinear Grey Model, ARIMA Model, NGM-ARIMA hybrid correction model