A SEASONAL FORECASTING MODEL OF THE TOURIST SECTOR: THE CASE OF CENTRAL MACEDONIA.

This paper presents a time series forecasting method to predict of occupancy of all tourist accommodations for the Central Macedonia region. Official statistical data is used from the year 1990 up to 1999. We analyze the trend and the seasonality, and construct statistical models, using the Box – Jenkins approach. After a series of appropriate statistical tests, we choose the model whish has the best fit to our data. The prediction ability of the model is judged by taking after forecast of the last twelve observations of our time series and comparing them with the real data. Finally we predict the percentages of occupancy of accommodation in the specific region for the next two years. From this study it is obvious that the occupancy of all tourist accommodations will increase during the low and the high tourist season for the next years. Keywords: Forecasting models, Time series, SARIMA