Multi-Horizon LSTM Forecasting of Indonesian Food Commodity Prices to Support the Free Nutritious Meal (MBG) Program

Deni Utama; Della Novita Sari

doi:10.34148/teknika.v15i1.1453

Authors

Deni Utama Department of Information Systems, Faculty of Computer Science, Universitas Negeri Jakarta, DKI Jakarta, Indonesia
Della Novita Sari Communication Studies, College of Communication, Fine Arts, and Media, University of Nebraska at Omaha, Nebraska, United States

DOI:

https://doi.org/10.34148/teknika.v15i1.1453

Keywords:

Long Short-Term Memory (LSTM), Multi-Horizon Forecasting, Food Commodity Prices, Free Nutritious Meal (MBG), Time Series Analysis

Abstract

This study develops a multi-horizon Long Short-Term Memory (LSTM) model to forecast food commodity prices in Indonesia, supporting the national Free Nutritious Meal (MBG) program. The MBG initiative increases daily and annual food demand for over 50 million beneficiaries, affecting rice, meat, eggs, vegetables, and cooking oil. Historical price data were used to train the model, which effectively captures nonlinear temporal patterns and seasonal variations. Forecast results indicate heterogeneous volatility patterns across commodities. Staple commodities such as rice and cooking oil exhibit relatively stable trends with minor monthly fluctuations, whereas perishable and climate-sensitive commodities such as chili, eggs, and sweet potatoes demonstrate higher short-term volatility. To ensure temporal robustness, the model was evaluated using a Rolling Forecast Origin (RFO) validation strategy with an expanding window approach. Evaluation on the test set yielded a MAE of 2,108.37, RMSE of 3,933.30, and MAPE of 6.61%, indicating approximately 93% accuracy. The model is implemented in an interactive interface allowing users to select commodities, set prediction horizons, and export results, providing a practical tool for procurement planning, distribution management, and national food security.

Downloads

Download data is not yet available.

References

[1] Badan Pangan Nasional, “Indeks Harga Pangan Nasional,” 2023.

[2] Sekretariat Negara Republik Indonesia, “Presiden dorong percepatan program Makan Bergizi Gratis di Indonesia,” 2025.

[3] Antara News, “Kepala Bapanas: Program Makan Bergizi Mampu Jadi Penggerak Ekonomi,” 2025.

[4] Food and Agriculture Organization, “The State of Food Security and Nutrition in the World,” 2023.

[5] A. J. S. a. B. Schölkopf, “A tutorial on support vector regression,” 2004.

[6] S. M. e. al., “The M4 Competition,” 2018.

[7] R. A. e. al., “Price volatility and food security in Indonesia,” 2022.

[8] H. Jayatri, “ARIMA–GARCH models for food prices,” 2025.

[9] S. a. F. Laili, “ARMA–GARCH for strategic food commodities,” 2023.

[10] D. U. e. al., “Public sentiment analysis on the MBG Program,” 2025.

[11] J. P. I. a. T. Handhayani, “ELM, LSTM, and ensemble methods,” 2025.

[12] S. H. a. J. Schmidhuber, “Long short-term memory,” 1997.

[13] A. S. a. M. Kotb, “Deep LSTM for petroleum forecasting,” 2019.

[14] S. B. e. al., “Multistep forecasting using LSTM,” 2020.

[15] B. A. C. e. al., “Hybrid forecasting of agricultural commodities,” 2025.

[16] R. e. al., “Optimization of average rice price forecasting in East Java using WFITS methods,” Jurnal Ekonomi dan Pembangunan (BRIN), 2019.

[17] R. L. M. e. al., “Enhancing agricultural commodity price forecasting with deep learning: A comparative study across 23 commodities,” Scientific Reports, 2025.

[18] F. T. a. K. D. R. Ly, “Forecasting commodity prices using long short-term memory neural networks,” arXiv preprint, 2021.

[19] Z. C. e. al., “Automated agriculture commodity price prediction system with machine learning techniques,” arXiv preprint, 2021.

[20] M.-K. G. e. al., “Forecasting commodity price shocks using temporal and semantic fusion of price signals and generative-AI extracted economic news,” arXiv preprint, 2025.