Peramalan Harga Minyak Goreng Menggunakan Model Deret Waktu Multivariat Dengan Variabel Eksogen (Study Case : Pulau Jawa)

Lamas, Valenitnus Pepinda (2026) Peramalan Harga Minyak Goreng Menggunakan Model Deret Waktu Multivariat Dengan Variabel Eksogen (Study Case : Pulau Jawa). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002211138-undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Minyak goreng merupakan salah satu komoditas strategis yang termasuk dalam kelompok sembilan bahan pokok dan memiliki peran penting bagi rumah tangga, UMKM, serta industri makanan. Dalam beberapa tahun terakhir, harga minyak goreng di Pulau Jawa mengalami fluktuasi yang signifikan dan berbeda antar provinsi, sehingga berdampak pada stabilitas ekonomi dan daya beli masyarakat. Salah satu faktor eksternal yang turut memengaruhi fluktuasi harga tersebut adalah nilai tukar Rupiah terhadap Dolar Amerika Serikat. Metode peramalan diperlukan menangkap keterkaitan antarwilayah sekaligus mempertimbangkan pengaruh variabel eksternal. Penelitian ini bertujuan untuk memodelkan dan meramalkan harga minyak goreng antarprovinsi di Pulau Jawa menggunakan model deret waktu multivariat VARMA dan VARMAX dengan variabel eksogen berupa nilai tukar Rupiah terhadap Dolar AS. Tahapan analisis meliputi uji stasioneritas data, identifikasi model, estimasi parameter menggunakan Kalman filter dan Maximum Likelihood Estimation, uji diagnostik residual, serta evaluasi kinerja model berdasarkan nilai Mean Absolute Percentage Error (MAPE).Hasil penelitian menunjukkan bahwa model VARIMAX mampu memberikan akurasi peramalan yang lebih baik dibandingkan model VARIMA, yang ditunjukkan oleh nilai MAPE yang lebih kecil pada data pengujian dengan masing-masing nilai MAPE pada Daerah Istimewa Yogyakarta sebesar 2,12%, DKI Jakarta 2,18%, dan Jawa Tengah sebesar 1,83%. Hal ini mengindikasikan bahwa penambahan variabel eksogen berupa nilai tukar Rupiah berperan penting dalam meningkatkan akurasi peramalan harga minyak goreng.
======================================================================================================================================
Cooking oil is one of the strategic commodities included in the nine basic necessities and plays an important role for households, MSMEs, and the food industry. In recent years, cooking oil prices in Java have experienced significant fluctuations and differences between provinces, thereby impacting economic stability and people's purchasing power. One of the external factors that has contributed to these price fluctuations is the exchange rate of the rupiah against the US dollar. Forecasting methods are needed to capture interregional correlations while considering the influence of external variables. This study aims to model and forecast cooking oil prices between provinces on the island of Java using the VARMA and VARMAX multivariate time series models with the exchange rate of the Rupiah against the US Dollar as an exogenous variable. The analysis stages include data stationarity testing, model identification, parameter estimation using the Kalman filter and Maximum Likelihood Estimation, residual diagnostic testing, and model performance evaluation based on the Mean Absolute Percentage Error (MAPE) value. The results show that the VARMAX model provides better forecasting accuracy than the VARMA model, as indicated by smaller MAPE values in the test data, with MAPE values of 2.12% for the Special Region of Yogyakarta, 2,18% for DKI Jakarta, and 1,83% for Central Java. This indicates that the addition of an exogenous variable in the form of the rupiah exchange rate plays an important role in improving the accuracy of cooking oil price forecasts.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	minyak goreng, VARMA, VARMAX, deret waktu multivariat, nilai tukar Rupiah cooking oil, VARMA, VARMAX, multivariate time series, Rupiah exchange rate
Subjects:	Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Valentinus Pepinda Lamas
Date Deposited:	27 Jan 2026 02:36
Last Modified:	27 Jan 2026 02:37
URI:	http://repository.its.ac.id/id/eprint/130478

Actions (login required)

View Item