Predicting Food Prices in Kenya using Machine Learning: A Hybrid Model Approach with XGBoost and Gradient Boosting

Benard O. Ogol¹Evans Omondi^1,2JOHN OLUKURU¹Betsy Muriithi¹Kennedy Senagi^1,3*

• ¹Strathmore University, Nairobi, Kenya
• ²African Population and Health Research Center, Nairobi, Kenya
• ³International Centre of Insect Physiology and Ecology (ICIPE), Nairobi, Kenya

Food price volatility continues to be a significant field of concern in Kenya's economic development and presents challenges to the stability of the country's economy. This study investigates the application of machine learning methods, using a hybrid approach of XGBoost and gradient boosting to help predict food prices in Kenya. The food prices data from the World Food Program from January 2006 to September 2024, currency exchange rates data from the Central Bank of Kenya in US dollars (USD), and inflation rates data were collated and pre-processed to be analytics and machine learning ready. The augmented data were preprocessed and transformed, then used to train XGBoost, gradient boosting, LightGBM, decision tree, random forest, and linear regression. A hybrid model was then developed by stacking XGBoost and gradient boosting as the base models, with linear regression serving as the meta-model used to combine their predictions. This model was then tuned by the hyperparameter random search method, achieving a mean absolute error of 0.1050, a mean squared error of 0.0261, a root mean square error of 0.1615, and an R-square of 0.9940, thereby surpassing the performances of all standalone models. We then applied cross-validation using 5–folds and Diebold–Mariano to check for model overfitting and model superiority analysis. Feature importance analysis using SHapley Additive exPlanations (SHAP) indicated that intuitive features impacting food prices are unit quantity, price type, commodity, and currency, while geographical factors like county have less impact. Finally, the model and the important features were then saved as pickle files to help deploy the model on a web application to facilitate food price predictions. This data-driven decision support system can help policymakers and agricultural stakeholders (such as the Kenyan government) plan for future trends in food prices. This can help prevent food insecurity in Kenya.