Modelling international tourism demand using seasonal ARIMA models

Purpose – The purpose of this study is to establish a seasonal autoregressive integrated moving average model able to capture and explain the patterns and the determinants of German tourism demand in Croatia. Design – The present study is based on the Box-Jenkins approach in building a seasonal autoregressive integrated moving average model intend to describe the behaviour of the German tourists’ flows to Croatia. Approach – The proposed model is a seasonal ARIMA(0,0,0)(1,1,3)4 model. Findings – The diagnostic checking and the performed tests showed that the estimated seasonal ARIMA(0,0,0)(1,1,3)4 model is adequate in modelling and analysing the number of German tourists ‘arrivals to Croatia. Originality of the paper – This study provides a seasonal ARIMA model helpful to analyse, understand and forecast German tourists’ flows to Croatia. Such, more detailed and systematic studies should be considered as starting points of future macroeconomic development strategies, pricing strategies and tourism sector routing strategies in Croatia, as a predominantly tourism oriented country