engleski

Forecasting Disaggregated Tourist Arrivals in Croatia: Evidence from Seasonal Univariate Time Series Models

This study examines the performance of four alternative univariate seasonal time series forecasting models (seasonal ARIMA, seasonal ARIMA with Fourier transformation, ARAR, and fractionally integrated ARMA) of tourist arrivals to twenty Croatian counties and the City of Zagreb. Both in-sample and out-of-sample forecasts reveal that the seasonal ARIMA model with Fourier transformation consistently outperforms the other models across the respective regions investigated.

tourist arrivals ; forecasting ; Croatia ; ARIMA ; ARAR ; Fourier transformation

Objavljeno prvi put na mreznim stranicama 1 srpnja 2015.