Statistical analysis of tourist arrivals to South Africa from non- Southern African Development Community African countries and the unspecified group, an under-tapped market

Abstract

Tourism is a sector that has gained significant attention from policy formulators and decision-makers around the globe, given its contribution to the overall development of the host country and beyond. The paper suggests a suitable Seasonal Autoregressive Integrated Moving Average (SARIMA) model to forecast non-Southern African Development Community (non-SADC) African countries and the Unspecified tourist arrivals to South Africa (SA). Tourist arrivals to SA from non-SADC African countries and those who did not specify country of origin (the Unspecified group) from July 2010 to February 2024 are examined. The non-SADC African countries comprise 46 countries, and the Unspecified group comprises those tourists who failed to specify their country of origin at their various ports of entry. A time series modelling approach is considered to support better-informed decision-making, forecasting, and assessment of the impact of the COVID-19 pandemic in the tourism sector. A SARIMA(2,1,0)(1,0,1)12 was considered the best model based on its lowest Akaike Information Criterion (AIC) value. Residual analysis is done to validate model accuracy and to support the fitted model. A graph of the forecasts and actuals highlights the negative impact of COVID-19 on tourist arrivals, the tourism sector and the country at large. There was a major step change in this category of tourist arrivals to SA due to the COVID-19 containment measures.