Forecasting international tourist flows to Australia: a comparison between the direct and indirect methods

This paper compares between the use of direct and indirect forecasting of international tourist flows to Australia. The indirect method is applicable when time series is a contemporaneous aggregate of others, which is often the case for tourist flows time series. In this case forecasts are generated by aggregating those of the component variables of aggregation. An alternative is the direct method, whereby forecasts are generated directly from the aggregated variable. This paper uses seasonal ARIMA models, regression-based models and Harvey's structural time series models to compare the accuracy of the two forecasting methods. We find evidence strongly in favour of the indirect method using monthly time series for international tourist arrivals to Australia.