Machine Learning-Based Forecasting Model for Paddy Rice Price movements Using Climatic and Macroeconomic Variables in Gombe State, Nigeria

Paddy rice is a major staple crop in Gombe State, Nigeria, contributing significantly to food security, household income, and rural livelihoods. However, paddy rice prices are highly volatile due to the combined effects of climatic variability, macroeconomic conditions, and market dynamics, creating uncertainty for farmers, traders, and policymakers. Existing forecasting studies have primarily focused on generalized agricultural commodities or relied on historical price data with limited integration of climatic and macroeconomic variables for localized price prediction. This study developed a machine learning-based forecasting model for predicting paddy rice price movements in Gombe State by integrating monthly historical paddy rice prices with climatic variables (rainfall, temperature, and relative humidity) and selected macroeconomic indicators obtained from the Climate Hazards Group InfraRed Precipitation with Station Data (CHIRPS), the Central Bank of Nigeria (CBN), the National Bureau of Statistics (NBS), FAOSTAT, and the Gombe State Agricultural Development Programme (GSADP). Five machine learning models, namely Random Forest, Extreme Gradient Boosting (XGBoost), LightGBM, CatBoost, and Long Short-Term Memory (LSTM), were developed and evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R²). The comparative evaluation showed that the Random Forest model achieved the best forecasting performance among the evaluated models. The developed forecasting framework provides a reliable decision-support tool for anticipating paddy rice price movements, reducing financial risk, improving production and marketing decisions, and supporting evidence-based agricultural policy formulation to enhance market efficiency in Gombe State, Nigeria.