Perbandingan Metode Prophet dan SARIMAX untuk Peramalan Harga Pangan di Lombok Barat

Aliva Nurramadhan; Dadang Priyanto; Neny Sulistianingsih

doi:10.31004/riggs.v4i4.4964

Authors

Aliva Nurramadhan Universitas Bumigora
Dadang Priyanto Universitas Bumigora
Neny Sulistianingsih Universitas Bumigora

DOI:

https://doi.org/10.31004/riggs.v4i4.4964

Keywords:

Harga Pangan, Peramalan, Prophet, SARIMAX, Deret Waktu

Abstract

Fluktuasi harga pangan di Kabupaten Lombok Barat merupakan permasalahan strategis yang berpengaruh terhadap stabilitas ekonomi daerah, kesejahteraan masyarakat, dan ketahanan pangan. Perubahan harga pangan dipengaruhi oleh tren jangka panjang, pola musiman, ketersediaan pasokan, serta dinamika permintaan. Kondisi ini menuntut metode peramalan yang mampu menghasilkan prediksi harga yang akurat dan andal sebagai dasar pengambilan keputusan dan perumusan kebijakan pangan daerah. Penelitian ini bertujuan membandingkan kinerja dua metode peramalan deret waktu, yaitu Prophet dan Seasonal Autoregressive Integrated Moving Average with Exogenous Regressors (SARIMAX), dalam memprediksi harga komoditas pangan utama di Kabupaten Lombok Barat. Data yang digunakan berupa data harga mingguan 13 komoditas pangan selama periode Januari 2021 hingga 5 September 2025 yang diperoleh dari Dinas Ketahanan Pangan Kabupaten Lombok Barat. Data diproses melalui tahapan praproses meliputi integrasi data, penambahan variabel waktu, normalisasi, serta pembentukan variabel eksogen berupa lag ketersediaan, kebutuhan, neraca, dan moving average harga. Pemodelan mempertimbangkan komponen tren, musiman, dan variabel eksternal. Evaluasi kinerja model menggunakan Mean Absolute Error (MAE) dan Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa secara rata-rata seluruh komoditas, model Prophet memiliki kinerja lebih baik dengan MAPE sebesar 3,80% dan MAE sebesar 1.398,25 dibandingkan SARIMAX dengan MAPE sebesar 4,89% dan MAE sebesar 1.524,68. Namun, secara per komoditas, Prophet unggul pada 7 komoditas, sedangkan SARIMAX lebih baik pada 6 komoditas lainnya. Temuan ini menunjukkan bahwa metode Prophet dan metode SARIMAX tidak menunjukkan keunggulan yang konsisten untuk seluruh komoditas pangan, sehingga pemilihan metode peramalan perlu disesuaikan dengan karakteristik data dan dinamika harga masing-masing komoditas.

Downloads

Download data is not yet available.

References

Badan Pusat Statistik Kabupaten Lombok Barat, “Produk Domestik Regional Bruto Kabupaten Lombok Barat Menurut Lapangan Usaha 2020-2024,” Lombok Barat, 2025.

S. Kwarteng and P. Andreevich, “Comparative Analysis of ARIMA, SARIMA and Prophet Model in Forecasting,” Research & Development, vol. 5, no. 4, pp. 110–120, 2024, doi: 10.11648/j.rd.20240504.13.

E. Kostaridou, N. Siatis, G. Lampiris, and E. Zafeiriou, “Empirical Comparison of Facebook Prophet and Traditional Models for Tomato Price Forecasting in Greece,” Research on World Agricultural Economy, vol. 5, no. 4, pp. 594–607, 2024, doi: 10.36956/rwae.v5i4.1295.

Tabatskova. olena, Kotykova. Mykola, Babych. Anna, Lagodzinska. Anna, “Agricultural and Resource Economics : International Scientific E-Journal Agricultural and Resource Economics : International Scientific E-Journal,” Agricultural and Resource Economics: International Scientific E -Journal, vol. 8, no. 2, pp. 30–49, 2022.

R. Yulianti, N. T. Amanda, K. A. Notodiputro, Y. Angraini, and L. N. A. Mualifah, “Comparison of Sarima and Sarimax Methods for Forecasting Harvested Dry Grain Prices in Indonesia,” Barekeng, vol. 19, no. 1, pp. 319–330, 2025, doi: 10.30598/barekengvol19iss1pp319-330.

Ms. S. R. S. Abirami, “Sales Prediction Based on Sarimax Time Series Algorithm,” International Scientific Journal of Engineering and Management, vol. 03, no. 03, pp. 1–9, 2024, doi: 10.55041/isjem01493.

X. Zhang, L. Zhang, and Z. Yang, “Comparison Between Elementary Network Structure and Prophet Model for Time-Series Forecasting: Merchandise Sales,” Highlights in Science, Engineering and Technology, vol. 101, pp. 104–115, 2024, doi: 10.54097/21h9k153.

M. D. Angelo, I. Fadhiilrahman, and Y. Purnama, “Comparative Analysis of ARIMA and Prophet Algorithms in Bitcoin Price Forecasting,” Procedia Comput Sci, vol. 227, pp. 490–499, 2023, doi: 10.1016/j.procs.2023.10.550.

A. Tripathi, A. Tripathi, O. Trenz, and P. Kumar Mishra, “Comparative Analysis of Selected Time Series Forecasting Approaches for Indian Markets,” pp. 167–186, 2024, doi: 10.11118/978-80-7509-990-7-0167.

J. Zhao and C. Zhang, “Research on sales forecast based on prophet-SARIMA combination model,” J Phys Conf Ser, vol. 1616, no. 1, 2020, doi: 10.1088/1742-6596/1616/1/012069.

R. Indriyati, A. Burhanuddin, and A. Utami, “Komparasi Sarimax dan Prophet untuk Prediksi Tingkat Penghunian Kamar Hotel Jawa Tengah Comparison Of Sarimax and Prophet for Predicting The Hotel Room Occupancy Rate In Central Java,” Techno.COM, vol. 23, no. 2, pp. 294–305, 2024.

A. L. M. Serrano et al., “Statistical Comparison of Time Series Models for Forecasting Brazilian Monthly Energy Demand Using Economic, Industrial, and Climatic Exogenous Variables,” Applied Sciences (Switzerland), vol. 14, no. 13, pp. 1–32, 2024, doi: 10.3390/app14135846.

D. Kunac, “Comparative analysis of time series models for river flood forecasting,” no. May, 2025, [Online]. Available: https://atrium.lib.uoguelph.ca/bitstreams/cc4e6c73-3e52-45fe-9f00-efab1b3e538c/download

Wang Heng, “Forecasting Egg Prices: A Time Series Analysis with Economic and Market Influences,” no. 2722303, Jan. 2025.

D. de S. B. Simão, “Meat consumption prediction: A data science perspective,” Lisbon School of Economics & Management, Universidade de Lisboa, 2023.

M. Dmytryshyn, “Development of a predictive model for sales forecasting of perishable food products,” pp. 1–27, 2023.

O. Khin and S. Lee, “Integrating Crop Types, Yield Products, and Price Forecasting with Explainable AI for Agricultural APIs,” pp. 0–28, 2025, doi: 10.20944/preprints202507.1164.v1.

L. Menculini et al., “Comparing Prophet and Deep Learning to ARIMA in Forecasting Wholesale Food Prices,” Forecasting, vol. 3, no. 3, pp. 644–662, 2021, doi: 10.3390/forecast3030040.

S. K. Majhi, R. Bano, R. Pradhan, and S. Panda, “Food Price Index Prediction Using Time Series Models: A Study of Cereals, Millets and Pulses,” pp. 216–238, 2025, doi: 10.1007/978-3-031-94997-5_21.

G. Saqware and I. Beary, “Hybrid SARIMA-Facebook Prophet Model for Prediction and Forecasting of the Staple Food Prices in Tanzania,” International Journal of Operations Research, vol. 20, pp. 1–11, 2023, [Online]. Available: http://doi.org/10.6886/IJOR.202308_20.0001

F. Ivan, А. Б. Кулик, О. О. Мельник, З. В. Сучасних, М. Аналізу, and Ч. Рядів, “Modeling and forecasting of gross milk yield using modern methods of time series analysis моделювання та прогнозування валового надою молока з використанням сучасних методів аналізу часових рядів,” pp. 247–252, 2025, doi: 10.20535/2307-5651.33.2025.335927.

T. Feng et al., “The comparative analysis of SARIMA, Facebook Prophet, and LSTM for road traffic injury prediction in Northeast China,” Front Public Health, vol. 10, 2022, doi: 10.3389/fpubh.2022.946563.

G. C. Lee, “A Data-Driven Approach to Tourism Demand Forecasting: Integrating Web Search Data into a SARIMAX Model,” Data (Basel), vol. 10, no. 5, 2025, doi: 10.3390/data10050073.

F. M. H. Atamimi, W. Wintanti, and G. Abdillah, “Enhanching Prophet Time Series Forecasting on Sparse Data via Hyperparameter Optimizattion: A Case Study in Retail,” Sinkron, vol. 9, no. 2, pp. 1000–1007, 2025, doi: 10.33395/sinkron.v9i2.14804.

M. M. Hoque, I. Ahmed, N. Banik, and M. M. Hoque, “Analyzing Price Forecasting of Grocery Products in Bangladesh: A Comparison of Time Series Modeling Approaches,” Lecture Notes in Networks and Systems, vol. 855 LNNS, no. December, pp. 334–341, 2023, doi: 10.1007/978-3-031-50158-6_33.