On modelling of development of tourism in selected European countries

Understanding tourism development over time requires analysing time series of available data and assessing relevant information and essential tourism indicators. The aim of this paper is modelling of the tourism development of the 24 European Union countries using Box-Jenkins methodology to capture and explain the patterns and the determinants of tourism in the European Union countries. The variable “Total Nights Spent at Tourist Accommodation Establishments” per month was considered, and was recorded for the period from January 2002 to September 2018 by the Eurostat database, since this is one of the variables that best expresses effective demand. Results of founded 24 SARIMA models reported in the overview table can be considered as preliminary analysis for next examining using ARFIMA models with long memory, artificial neural networks, or models based on Engle's methodology. For the best six models, ex-ante analysis of accuracy is done. The best model among them is the model for Germany.