Unemployment rate and the gross domestic product in Somalia: Using frequentist and Bayesian approach

Abstract

Gross domestic product (GDP) serves as a vital indicator of a country’s economic health, influenced by factors such as unemployment rates, export-import dynamics, and inflation rates. Understanding the intricate relationship between GDP and unemployment is crucial for navigating the dynamic business environment. This study investigates this relationship in Somalia using both frequentist and Bayesian regression approaches. The frequentist approach, particularly ordinary least squares, is chosen for its widespread use and simplicity in estimating parameters by minimizing the sum of squared deviations. Meanwhile, the Bayesian approach is adopted for its flexibility in integrating prior information and updating estimates as new data emerge, providing a probabilistic framework that accommodates parameter uncertainty. Secondary data from the World Bank spanning 1991–2020 were analyzed. The results reveal a significant negative association between unemployment and GDP in both models. Comparative analysis employing mean absolute error, root mean square error, and mean square error indicates comparable predictive accuracy between the two approaches, with the Bayesian model demonstrating slight advantages. Bayesian convergence diagnostics affirm satisfactory model stability. These findings offer valuable insights for policymakers, aiding in the formulation of strategies to address unemployment challenges and stimulate economic growth in Somalia.

Keywords

• GDP
• unemployment rate
• Bayesian modeling
• regression linear model
• Somalia
• Agricultural Economics