Analisis Komparasi Kinerja ARIMA dan DES Holt Menggunakan Multi-Split Validation pada Peramalan Harga BBM Indonesia
DOI:
https://doi.org/10.62712/juktisi.v5i2.1367Keywords:
Peramalan Harga BBM, ARIMA, Double Exponential Smoothing Holt, Multi-Split Validation, Time Series, Mean Absolute Percentage ErrorAbstract
Harga bahan bakar minyak (BBM) di Indonesia sangat rentan terhadap dinamika pasar energi global, sehingga kemampuan peramalan yang akurat menjadi instrumen penting dalam perencanaan kebijakan energi nasional. Penelitian ini bertujuan membandingkan kinerja dua metode peramalan deret waktu, yaitu Double Exponential Smoothing (DES) Holt dan AutoRegressive Integrated Moving Average (ARIMA), dalam memprediksi harga bensin di Indonesia menggunakan pendekatan multi-split validation. Dataset bersumber dari Kaggle (World vs Asia Fuel Prices) mencakup 136 data bulanan dari Januari 2015 hingga April 2026. Validasi dilakukan pada tiga skenario rasio pembagian data (70:30, 80:20, dan 90:10) guna memverifikasi kekokohan (robustness) setiap model. Pemilihan orde optimal ARIMA dilakukan melalui grid search berbasis AIC, menghasilkan konfigurasi ARIMA(1,1,0). Kinerja model dievaluasi menggunakan metrik MAE, MSE, RMSE, dan MAPE. Hasil penelitian menunjukkan bahwa ARIMA(1,1,0) secara konsisten unggul dibandingkan DES Holt pada seluruh skenario. Pada rasio 80:20, ARIMA meraih MAPE 13,94% dibanding DES Holt 16,75%, keduanya masuk kategori "Akurat". Temuan ini membuktikan bahwa pola autokorelasi kompleks pada data harga BBM Indonesia lebih efektif dimodelkan oleh ARIMA dibandingkan asumsi tren linier pada DES Holt.
Downloads
References
[1] M. G. Gamary, “Peran Penting APBN dalam menjaga Stabilitas Harga BBM,” Mentri Keuangan Republik Indonesia Lubuk Sikaping. Accessed: Jun. 22, 2026. [Online]. Available: https://djpb.kemenkeu.go.id/kppn/lubuksikaping/id/data-publikasi/artikel/3281-peran-penting-apbn-dalam-menjaga-stabilitas-harga-bbm.html
[2] J. Baek, “Crude oil prices and macroeconomic activities: a structural VAR approach to Indonesia,” Appl. Econ., vol. 53, no. 22, pp. 2527–2538, May 2021, doi: 10.1080/00036846.2020.1862750.
[3] M. S. Priyadani and R. I. S. Setiawati, “FROM EXPORTER TO IMPORTER: TRACING THE SHIFTING DYNAMICS OF INDONESIA’S OIL TRADE PERIOD 2001-2022,” JEA17 J. Ekon. Akunt., vol. 10, no. 2, pp. 1–15, Oct. 2025, doi: 10.30996/JEA17.V10I2.132122.
[4] Q. Zhang, Y. Hu, J. Jiao, and S. Wang, “The impact of Russia–Ukraine war on crude oil prices: an EMC framework,” Humanit. Soc. Sci. Commun. 2024 111, vol. 11, no. 1, pp. 8-, Jan. 2024, doi: 10.1057/s41599-023-02526-9.
[5] C. Gharib, S. Mefteh-Wali, V. Serret, and S. Ben Jabeur, “Impact of COVID-19 pandemic on crude oil prices: Evidence from Econophysics approach,” Resour. Policy, vol. 74, p. 102392, Dec. 2021, doi: 10.1016/J.RESOURPOL.2021.102392.
[6] M. As’ad, E. Yuniar, Sujito, E. Farida, and S. Setyowibowo, “Crude Oil Price Forecasting with Double Exponential Smoothing-Holts Model,” G-Tech J. Teknol. Terap., vol. 7, no. 1, pp. 201–208, Jan. 2023, doi: 10.33379/GTECH.V7I1.1901.
[7] M. Latif, E. Arfian, P. Pardede, M. Sipahutar, and P. Naibaho, “Forecasting Stock Prices of PT. Bank Negara Indonesia (Persero) Tbk., by Method (BOX-JENKINS),” Primanomics J. Ekon. Bisnis, vol. 19, no. 1, pp. 191–205, Jan. 2021, doi: 10.31253/PE.V19I1.520.
[8] N. Andriani, S. Wahyuningsih, and M. Siringoringo, “Application of Double Exponential Smoothing Holt and Triple Exponential Smoothing Holt-Winter with Golden Section Optimization to Forecast Export Value of East Borneo Province,” J. Mat. Stat. dan Komputasi, vol. 18, no. 3, pp. 475–483, May 2022, doi: 10.20956/J.V18I3.17492.
[9] M. Z. Y. Nurul Azizah Muzakir, “Analisis Perbandingan Model Double Exponential Smoothing dan ARIMA untuk Prediksi Harga Beras di Indonesia,” J. Stat. Its Appl. Teach. Res., vol. 7, no. 1, pp. 7–20, 2025, doi: 10.35580/variansiunm.
[10] D. J. Pedregal, “New algorithms for automatic modelling and forecasting of decision support systems,” Decis. Support Syst., vol. 148, p. 113585, Sep. 2021, doi: 10.1016/J.DSS.2021.113585.
[11] A. K. Wicaksono, T. Prasetyo, and N. Az, “STORAGE SERVER DATABASE UTILIZATION FORECASTING USING HOLT-WINTERS AND ARIMA METHODS IN E-GOVERNMENT SYSTEM. STUDY AT KEMENKEU RI,” J. Tek. Inform., vol. 4, no. 6, pp. 1399–1408, Sep. 2023, doi: 10.52436/1.JUTIF.2023.4.6.1147.
[12] T. M. A. Khuram Shahzad, “World vs Asia Fuel Prices,” 2026. [Online]. Available: https://www.kaggle.com/datasets/zkskhurram/world-vs-asia-fuel-prices
[13] D. A. I. Asrul and A. A. Soebroto, “Optimalisasi Prediksi Kasus COVID-19 di Indonesia: Perbandingan Teknik Validasi 80-20 Split dan Walk-Forward dengan ARIMA,” J-INTECH, vol. 12, no. 02, pp. 297–308, Dec. 2024, doi: 10.32664/J-INTECH.V12I02.1373.
[14] P. D. A. Andini, S. Wahyuningsih, and M. Siringoringo, “Aplikasi Metode Double Exponential Smoothing Holt Dengan Optimasi Golden Section Untuk Peramalan Nilai Ekspor Provinsi Kalimantan Timur,” EKSPONENSIAL, vol. 15, no. 1, pp. 20–28, May 2024, doi: 10.30872/EKSPONENSIAL.V15I1.1278.
[15] Y. J. Mgale, Y. Yan, and S. Timothy, “A Comparative Study of ARIMA and Holt-Winters Exponential Smoothing Models for Rice Price Forecasting in Tanzania,” OALib, vol. 08, no. 05, pp. 1–9, 2021, doi: 10.4236/OALIB.1107381.
[16] R. Syahril Amanu et al., “PERBANDINGAN MODEL PREDIKSI DATA MINING DALAM MEMPREDIKSI KONSENTRASI POLUTAN KARBON MONOKSIDA (CO) DI JAKARTA,” J. Teknol. Inf. J. Keilmuan dan Apl. Bid. Tek. Inform., vol. 18, no. 1, pp. 7–21, Jan. 2024, doi: 10.47111/JTI.V18I1.12451.
[17] Y. N. Hilal, G. D. A. Nainggolan, S. H. Syahputri, and F. Kartiasih, “Comparison of ARIMA and LSTM Methods in Predicting Jakarta Sea Level,” J. Ilmu dan Teknol. Kelaut. Trop., vol. 16, no. 2, pp. 163–178, Aug. 2024, doi: 10.29244/JITKT.V16I2.52818.
[18] S. Saniuk et al., “The Impact of COVID-19 and War in Ukraine on Energy Prices of Oil and Natural Gas,” Sustain. 2023, Vol. 15, Page 14208, vol. 15, no. 19, p. 14208, Sep. 2023, doi: 10.3390/SU151914208.
[19] S. Suliadi, “Kode R dan Selang Kepercayaan Korelasi Berdasarkan Empirical Likelihood serta Implementasinya pada Korelasi PDRB dengan Jumlah Kasus Covid-19 di Indonesia,” Statistika, vol. 22, no. 1, pp. 1–11, Sep. 2022, doi: 10.29313/STATISTIKA.V22I1.357.
[20] J. Zhang, Y. Yang, and J. Ding, “Information criteria for model selection,” Wiley Interdiscip. Rev. Comput. Stat., vol. 15, no. 5, p. e1607, Sep. 2023, doi: 10.1002/WICS.1607;PAGE:STRING:ARTICLE/CHAPTER.
[21] H. Alabdulrazzaq, M. N. Alenezi, Y. Rawajfih, B. A. Alghannam, A. A. Al-Hassan, and F. S. Al-Anzi, “On the accuracy of ARIMA based prediction of COVID-19 spread,” Results Phys., vol. 27, p. 104509, Aug. 2021, doi: 10.1016/J.RINP.2021.104509.
[22] V. R. Joseph, “Optimal ratio for data splitting,” Stat. Anal. Data Min., vol. 15, no. 4, pp. 531–538, Aug. 2022, doi: 10.1002/SAM.11583;SUBPAGE:STRING:FULL.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Alpon Siyus, Wahyu Nugraha, Rabiatus Saadah
This work is licensed under a Creative Commons Attribution 4.0 International License.
