Modeling and forecasting the impact of climatic and non-climatic factors on banana production in Somalia: A comprehensive analysis of the somali agricultural sector

Abstract

This study explores the impact of non-climatic and climatic factors on banana production in Somalia, using time series data from 1961 to 2018. The banana industry in Somalia has held a crucial role in the country's exports, ranking as the second most important sector, following livestock. Somalia has maintained its position as the leading banana producer in East Africa for over a century, but the market stalled due to the 1991 civil war. This study aims to examine the impact of climatic and non-climatic factors on Banana production in Somalia through modeling and projection. The study used 63-year annual time series data from secondary sources from the FAO Centre and the World Bank. It used Vector Autoregressive and Autoregressive Distributive Lag models to analyse variables' stationary behaviour. Unit root tests were applied, and the optimal lag order was determined using minimizing information criteria or maximizing log-likelihood. Results showed that most variables became stationary after the first difference, except for CO2 emissions and urban population. The researcher recommends using an autoregressive disturbed lag model instead of a Vector Autoregressive model. Long-term, temperature and political stability have a significant positive impact on the dependent variable. At the same time, rainfall, CO2 emissions, land under cereal, and urban population are not statistically significant predictors in the short run. The study confirms a long-term equilibrium connection between climate variables, non-climate variables, and agricultural output. Climate factors, particularly temperature, positively influence banana output, while non-climatic factors like cultivated land and political stability have a positive impact. These findings offer guidance for policymakers in designing effective interventions.