Time Series Forecasting of Beef Production in Somalia Using Single and Hybrid Models

Background Beef production is a critical component of Somalia’s livestock sector, contributing substantially to food security, rural livelihoods, export earnings, and national economic development. Reliable forecasting of beef production is essential for evidence-based planning and policy formulation, particularly in a climate-vulnerable and drought-prone environment. Methods This study applied and compared a range of single and hybrid time-series forecasting models using annual beef and buffalo meat production data for Somalia from 1961 to 2023. Single models included Autoregressive Integrated Moving Average (ARIMA), Exponential Smoothing State Space (ETS), Theta, Trigonometric Seasonality, Box-Cox Transformation, ARMA Errors, Trend and Seasonal Components (TBATS), Autoregressive Fractionally Integrated Moving Average (ARFIMA), and Neural Network Autoregression (NNAR). Hybrid models were constructed by combining selected forecasting approaches. Model performance was evaluated using Mean Absolute Percentage Error (MAPE), Symmetric Mean Absolute Percentage Error (sMAPE), and Theil’s U statistic. Results The results demonstrated substantial variation in forecasting performance across models. Among the single models, ARFIMA achieved the highest predictive accuracy, yielding the lowest error values (MAPE = 2.76%, sMAPE = 0.0271, Theil’s U = 2.41). Among the hybrid approaches, ARIMA–ETS performed best, although it did not outperform the ARFIMA model. Forecasts generated by ARFIMA suggest a modest decline in annual beef production from approximately 52,631 metric tonnes in 2024 to 51,573 metric tonnes in 2030. Prediction intervals widened over time, indicating increasing uncertainty associated with longer forecast horizons. Conclusions The findings indicate that long-memory dynamics play an important role in Somalia’s beef production system, making ARFIMA the most suitable forecasting model among those evaluated. Integrating quantitative forecasting tools into livestock planning frameworks may strengthen drought preparedness, production management, and evidence-based policy development. This study provides the first comprehensive comparative forecasting framework for beef production in Somalia and offers valuable insights for sustainable livestock sector planning.