Forecasting the Global Price of Corn Using Neural Network and Hybrid Approach

The global price of corn is critical in agricultural markets, food security, and trade policies, making accurate forecasting essential for policymakers and stakeholders. Traditional econometric forecasting models have been widely used but often fail to capture market volatility, nonlinear trends, and external shocks, necessitating advanced machine-learning techniques. This study explored forecasting models, including traditional and advanced machine learning and hybrid approaches, to determine the most effective method for predicting corn prices. Using monthly corn price data from 2014 to 2024, the study evaluated model performance based on different evaluation metrics. The results indicated that the Neural Network Auto Regressive (NNETAR) model outperformed traditional econometric models, effectively capturing complex temporal structures and adapting to sudden market fluctuations. Additionally, hybrid models integrating NNETAR and Seasonal and Trend decomposition using LOESS Model (STLM) demonstrated further improvements in forecasting accuracy, suggesting that combining statistical and machine learning methods enhances predictive reliability. These findings underscore the potential of machine learning in agricultural price forecasting, providing valuable insights for policymakers to implement market stabilization strategies, adjust trade policies, and enhance risk management in the U.S. corn sector.