Forecasting Urban Population Growth in Somalia: Using ARIMA, TBATS, NNAR, and Hybrid Models

Rapid urbanization presents opportunities and challenges, particularly in developing nations like Somalia. Accurate forecasting of urban population growth is essential for effective urban planning and resource management. This study employs ARIMA, TBATS, NNAR, and hybrid models to forecast urban population growth in Somalia, utilizing historical data from 1961 to 2022 sourced from the World Bank. The analysis evaluates the performance of single models and hybrid combinations (ARIMA-TBATS, ARIMA-NNAR, ARIMA-TBATS-NNAR) based on metrics including Theil’s U statistic, MAPE, SMAPE, and RMSE. Results indicate that the TBATS model best fits among the single time-series models, while the ARIMA-TBATS-NNAR hybrid model outperforms the others in forecasting urban population growth. The validated models effectively predict an increase in urban residents in Somalia from 2022 to 2033. This study underscores the importance of leveraging advanced statistical modeling, particularly hybrid approaches, to inform evidence-based strategies and optimize resource allocation for sustainable urban development in Somalia, contributing to achieving Sustainable Development Goals (SDGs) 11 and 7 and instilling hope for a more sustainable future.