Claim Missing Document
Check
Articles

Found 1 Documents
Search

SELECTING THE BEST MODEL FOR FORECASTING INDONESIA'S OIL AND GAS IMPORT VALUE USING ARIMAX AND ARIMAX-LSTM Zega, Alvandi Syukur Rahmat; Hidayat, Anang Kurnia; Jannah, Nazwa Thoriqul; Kartiasih, Fitri
Dynamic Management Journal Vol 8, No 4 (2024): October
Publisher : Universitas Muhammadiyah Tangerang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31000/dmj.v8i4.10776

Abstract

In order for the government to make the best policy decisions going forward, it is critical to forecast the value of oil and gas imports. This study aims to anticipate Indonesia's oil and gas import value by taking into account the independent variables of inflation, rupiah exchange rates, and global crude oil prices. The ARIMA (Autoregressive Integrated Moving Average), ARIMAX (ARIMA With Exogenous Variable), and Hybrid ARIMAX-LSTM (ARIMA With Exogenous Variable Long-Short-Term Memory) are the methods that are compared. Mean Absolute Percentage Error, or MAPE, is a tool used to compare forecasting models. The outcomes demonstrated how well ARIMAX-LSTM (0, 1, 2) predicts and forecasts the value of oil and gas imports when combined with variables for inflation and crude oil prices. According to the forecasting results, the value of imports of gas and oil increased by 3.03% between January and September of 2024 when compared to the entire import value of the year prior (Y-o-Y). Other exogenous variable addition, additional research on hyperparameter tuning, and the use of cross-validation techniques to increase prediction accuracy and provide more precise measurements of model performance are other recommendations for future investigation.