Analysis and Forecasting Prices and Volatility of Brent Oil Using the ARIMA-GARCH Model

Source abstract

Global oil prices, particularly Brent Crude Oil, experience high fluctuations and are influenced by various economic and geopolitical factors. This situation emphasizes the importance of volatility analysis and oil price forecasting as a basis for economic decision-making. This study aims to analyze the price characteristics and volatility of Brent oil for the 2020–2025 period, determine the best forecasting model using ARIMA–GARCH, and evaluate the accuracy of the forecasting model. The data used are daily Brent oil closing price data obtained from Yahoo Finance. The analytical methods used include descriptive statistics, stationarity tests, ACF and PACF analysis, ARIMA modeling, diagnostic tests, and GARCH modeling. The results indicate that Brent oil price data exhibits high volatility and contains ARCH effects, thus requiring a GARCH model to model data variance. The best model obtained is ARIMA(2,1,0)–GARCH(1,1). The forecasting results indicate that Brent oil prices tend to be stable in the short term, while volatility is predicted to remain persistent. The model accuracy evaluation yielded a MAPE value of 1.84% and an RMSE of 1.475, indicating excellent forecasting accuracy. The results of this study are expected to serve as a reference for investors and governments in understanding global oil price dynamics and economic decision-making.