International tourist arrivals modelling and forecasting: A case of Zimbabwe

Abstract

Zimbabwe is blessed with tourist attractions that draw visitors from all over the world. However, there are no quantitative models available for tourism stakeholders to utilize in decision-making and planning. The country is experiencing foreign currency shortages, which may be alleviated if the tourism industry, which has the power to generate foreign currency, adopted quantitative forecasting techniques that can provide reliable estimates. For planning reasons, resource mobilization, and allocation, accurate tourist projections are critical to the government and other tourism stakeholders. The goal of this research is to model international tourist arrivals in Zimbabwe and develop a quantitative statistical model that can be used to forecast future international tourist visitors. The Zimbabwe National Statistics Agency (ZIMSTAT) provided monthly foreign tourist arrivals data for the period January 2000 to December 2018. After the data revealed non-stationarity and seasonality, a time series technique in the form of the Box-Jenkins approach is applied to the data. The autocorrelation function (ACF), partial autocorrelation function (PACF), and root mean square error (RMSE) revealed that a seasonal autoregressive integrated moving average (SARIMA) model suited well to the data. The model predicted a gradual and seasonal increase in international tourist arrivals. The results of this model could be used by those in charge of tourism marketing to develop effective and efficient marketing strategies so that the country can receive a significant increase in international tourists, which will bring in much-needed foreign currency. It is important for tourism stakeholders and service providers to guarantee the availability of enough transport and accommodation facilities, especially during peak seasons.

Keywords

• International tourist arrivals
• Forecasting
• ARIMA
• SARIMA