MODEL PERGERAKAN HARGA MINYAK MENTAH BRENT MENGGUNAKAN PENDEKATAN TIME SERIES DENGAN EFEK LONG MEMORY

Abstract

Oil price movements are highly volatile and tend to be influenced over extended periods, often displaying long memory effect. This study utilizes the Autoregressive Fractionally Integrated Moving Average (ARFIMA) model, a long memory model, to analyze and forecast crude oil prices using monthly Brent data from November 1998 to November 2023. The analysis confirms the presence of long memory in the Brent crude oil price data. The ARFIMA model is then developed by estimating the parameter using the Rescaled Range Statistics (R/S) method. The best model is selected based on the lowest AIC and BIC values. The results indicate that the optimal ARFIMA model closely aligns with the actual data patterns, demonstrating its effectiveness in capturing the movements of Brent crude oil prices.

Keywords

• long memory, arfima, brent crude oil, rescaled range statistics (r/s).
• minyak mentah brent