FORECASTING INTERNATIONAL QUARTERLY TOURIST FLOWS USING ERROR-CORRECTION AND TIME-SERIES MODELS

This paper compares a range of forecasting models in the context of pr edicting quarterly tourist flows into Australia from the major tourist markets of USA, Japan, UK and New Zealand. Models considered include the error-correction model, the autoregressive model, the autoregressi ve integrated moving average model, the basic structural model and a r egression based time series model. Seasonality is an important feature of these series that requires careful handling. The relative performa nce of each model varies from country to country. The main conclusion is that relative to the time-series models, the error correction model s perform poorly. This may be caused by the way in which decisions on how best to model nonstationarity and seasonality are made. (C) 1997 E lsevier Science B.V.