Seasonal autoregressive integrated moving average time series model for tourism demand: the case of Sal Island, Cape Verde

Abstract

This article appears as an essential contribution for decision-makers in the Cape Verdean tourism sector given the impact that the number of overnight stays has on the economy of the country and the Sal Island, which until 2018 had been increasing every year. Since seasonality is a strong feature of the island's tourism, decision-makers are interested in knowing the seasonal variation in tourism demand. Thus, this study focused on the application of the Box-Jenkins method to the time series of the monthly number of nights stays in tourist establishments on the Sal Island, Cape Verde, over the period from January 2000 to December 2018, to find a model that better describes the series and with good forecast results for the year 2019. Several SARIMA models were studied using the Box-Jenkins method, with the SARIMA(1,1,1)(0,1,1)12 and the SARIMA(2,1,0)(0,1,1)12 demonstrating the best predictive performance in the test phase. However, in forecasting the series for the year 2019 the SARIMA(2,1,0)(0,1,1)12 achieved the best results with a MAPE=8.78%. This model can be used to simulate and analyse the number of overnight stays that be expected on the Island, if the tourism sector were not affected by the pandemic caused by COVID-19.