Modelling and forecasting Zimbabwe’s immigrants using SARIMA models.

Abstract

The remarkable increase in the Zimbabwe’s migration is of major concern in terms of planning purposes and policy formulation by the government, Non-governmental organizations (NGOs), tourism authorities, investors and industrialists. It is important to have statistical models that can be used to project future migrants as the projections are vital in decision making purposes. The total monthly migrants (immigrants) of Zimbabwe from 2001 to 2017 are modelled using the Box Jenkins technique. The data for the study was availed by the Zimbabwe National Statistics Agency (ZIMSTAT). A time series plot determined the migration trend to Zimbabwe and December is noted as the seasonal month with a significant number of migrants. The parsimonious SARIMA (1, 1, 1)(2,0,1)12 model fitted well to the migration data is validated by the Akaike information criterion (AIC), mean square error (MSE), root mean square error (RMSE) and other graphical measures. The model is used to project immigrants for the next 24 months. The projected migrants indicate a slow increase with the month of December being the peak month. The government of Zimbabwe, NGOs, investors and the industrialists should anticipate for an increased population due to an increase in immigrants and plan accordingly.

Keywords

• Migration
• immigrants
• ARIMA
• SARIMA
• forecasting