engleski

Dealing with seasonality: Modelling tourism demand in Croatia

The aim of the paper is to analyse tourism demand in Croatia and emphasise its seasonal patterns. The Croatian tourism is pronouncedly seasonal sensitive. Due to this fact, in modelling Croatian tourism demand, is necessary to implement quantitative methods that capture its seasonal character. Among the methods that can be used in analysing seasonal data, in this study the ratio-to-moving average method is used. After isolating the seasonal component of tourism demand several extrapolative methods were used with the purpose to model the number of tourist arrivals in Croatia ; Naïve 1, Naïve 2, single moving average, weighted moving average, single exponential smoothing, Browns double exponential smoothing and the autoregressive method. The monthly tourist arrivals in Croatia in the period from January 2003 to December 2013 are used as a commonly measure of tourism demand. The analysed data indicate the seasonal character of Croatian tourism demand. Taking this into account, the authors first isolated the seasonal component and emphasised the residual effects on the movement of the tourist demand. The-ratio-to-moving-average method is explained in detail and implemented to tourist arrivals data obtained from the secondary data sources. After modelling the seasonally adjusted values of tourist arrivals, chosen methods were evaluated by forecasting accuracy measure, the mean absolute percentage error – MAPE. Finally, at the end of the paper, the authors point out the importance of analysing seasonal phenomena, such as tourism demand, with the appropriate methods which take into account the seasonal component of the phenomenon itself.

croatian tourism ; ratio-to-moving-average ; seasonality ; tourism demand modelling ; extrapolative methods ; MAPE

nije evidentirano