Manuscript received December 18, 2025; accepted January 15, 2026; published January 27, 2026
Abstract—Accurate price forecasting is essential for informed decision-making by stakeholders in the agricultural sector, including farmers, traders, and policymakers. This study explored the use of deep learning models to predict market prices of four major agricultural commodities in Thailand, including rice, corn, cassava, and sugarcane. We evaluated and compared three network architectures across multiple forecasting horizons on a dataset from the Thai market. Overall, our findings suggested that the Long- and Short-term Time-series network or LSTNet was the most stable model, highlighting the advantage of capturing relationships among multiple variables. The results essentially pinpointed the strengths and limitations of each model, emphasizing the need for careful model selection based on characteristics of individual agricultural products.
Keywords—agricultural price forecasting, time series prediction, deep learning, recurrent neural networks, long short-term memory
Cite:Isara Chaowuttisuk and Papis Wongchaisuwat, "Forecasting Thai Agricultural Price: A Deep Learning Approach for Key Commodities," International Journal of Machine Learning vol. 16, no. 1, pp. 1-6, 2026.
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).