The Promise of Time-Series Foundation Models for Agricultural Forecasting: Evidence from Commodity Prices

2026-01-10

AI Summary

Agricultural market price forecasting has long been hindered by nonlinear dynamics, structural breaks, and data sparsity, often preventing complex models from outperforming traditional time series methods. This study systematically evaluates 17 approaches on USDA monthly price data spanning 1997–2025 and demonstrates, for the first time, that zero-shot time series foundation models—such as Time-MoE—achieve robust, high-accuracy predictions using only historical prices, without requiring exogenous covariates. Experimental results show that Time-MoE reduces overall mean absolute error (MAE) by 45% compared to the USDA benchmark and other machine learning models, with prediction errors for annual average prices of corn and soybeans decreasing by more than 50%. These findings significantly advance the paradigm of agricultural price forecasting toward pure time series modeling, eliminating reliance on external features.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Cognitive Modeling & Cognitive Systems: Other Foundations of Cognitive Modeling & Systems
• Computer Vision: Diffusion Models for Vision

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Forecasting agricultural markets remains challenging due to nonlinear dynamics, structural breaks, and sparse data. A long-standing belief holds that simple time-series methods outperform more advanced alternatives. This paper provides the first systematic evidence that this belief no longer holds with modern time-series foundation models (TSFMs). Using USDA ERS monthly commodity price data from 1997-2025, we evaluate 17 forecasting approaches across four model classes, including traditional time-series, machine learning, deep learning, and five state-of-the-art TSFMs (Chronos, Chronos-2, TimesFM 2.5, Time-MoE, Moirai-2), and construct annual marketing year price predictions to compare with USDA's futures-based season-average price (SAP) forecasts. We show that zero-shot foundation models consistently outperform traditional time-series methods, machine learning, and deep learning architectures trained from scratch in both monthly and annual forecasting. Furthermore, foundation models remarkably outperform USDA's futures-based forecasts on three of four major commodities despite USDA's information advantage from forward-looking futures markets. Time-MoE delivers the largest accuracy gains, achieving 54.9% improvement on wheat and 18.5% improvement on corn relative to USDA ERS benchmarks on recent data (2017-2024 excluding COVID). These results point to a paradigm shift in agricultural forecasting.

Problem

• agricultural forecasting
• time-series foundation models
• Market Year Average prices
• nonlinear dynamics
• structural breaks

Innovation

• Time-Series Foundation Models
• Zero-shot Forecasting
• Agricultural Price Prediction
• Time-MoE
• Market Year Average