Forecasting Wheat Prices in the United States Using an ARIMA Model
Bonny Amin
Department of Agricultural Sciences, Texas State University, San Marcos, United States.
Hurunnahar Khushi
*
Department of Agribusiness and Marketing, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh and Department of Environmental and Natural Resource Economics, The University of Rhode Island, Kingston, United States of America.
*Author to whom correspondence should be addressed.
Abstract
Backgrounds and Aims: Wheat is the third most produced field crops in the United States after corn and soybeans with an estimated 1.9 billion bushels produced in 2024/2025. As a leading global exporter, the U.S. exports approximately 50% of its total wheat production to international markets, making wheat price stability crucial for both domestic and global food security. This study aims to forecast future wheat prices in the United States using time-series ARIMA modelling to support farmers and traders in decision-making. The forecasts are intended to guide sowing, land allocation, and marketing decisions by providing price expectations ahead of planting seasons in the USA.
Study Design: Econometric time-series forecasting study using an Autoregressive Integrated Moving Average (ARIMA) model (Box–Jenkins methodology).
Place and Duration of Study: United States; the analysis covers annual wheat price data from 1991 to 2022.
Methodology: Yearly wheat price data were analyzed following the Box–Jenkins ARIMA framework. Stationarity was evaluated with graphical analysis and unit root tests, and the series was differenced to achieve stationarity. Autocorrelation and partial autocorrelation functions guided model identification. Competing ARIMA specifications were assessed based on log-likelihood, SigmaSQ (error variance), Akaike and Bayesian information criteria, and 95% confidence intervals. Model parameters were estimated using STATA software.
Results: The ARIMA (2,2,2) model provided the best fit for the data. Forecasts indicate a steady upward trend in U.S. wheat prices from 2024 through 2032. Predicted prices rise from 322.39 USD/ton in 2024 to 338.46 USD/ton in 2025 (+0.5% from 2024), reaching 455.44 USD/ton by 2032 (approximately 41% higher than the 2024 value).
Conclusion: Wheat prices in the United States are projected to increase consistently over the next decade. These projections offer valuable insights for farmers and distributors, aiding in informed sowing, land allocation, and marketing strategies.
Keywords: ARIMA, box–Jenkins methodology, wheat price forecasting, time series analysis, autocorrelation function, United States