Abstract
We forecast demand for Australian passports using a number of univariate and multivariate forecasting models, and assess their relative predictive ability over a number of forecasting horizons and evaluation measures. Our key result is to use different forecasting models for predicting passport demand in the short- versus medium- to long-run. Specifically, to forecast Adult-and-Senior passport demand in the short-term (i.e. up to 12 months) univariate ARIMA models are preferred, while for the longer term forecasts multivariate models with exogenous variables outperform, although only marginally. To forecast passport demand for Minors (less than 18 years old) ARIMA models perform well both in the short-term and the long-term, although ARIMA with explanatory variables outperforms slightly.
Acknowledgement
We thank Jingyuan Meng for excellent research assistantship. The initial work for this paper was commissioned by the Australian Passport Office, Department of Foreign Affairs and Trade.
Notes
This is mainly due to the fact that this activity has been conducted either in-house by various government departments or contracted out to third party private consultancies.
In some cases where certain explanatory variables become available after January 1987 we shift the beginning of the estimation period as required by the data, but keep the end of the in-sample interval fixed.
Seasonal uni-root test tables are available upon request. We verify the validity of our conclusions by checking the forecasting accuracy of models built on the seasonally differenced series and find that forecasting accuracy worsens.
A congruent model is the one that matches the data evidence on all measured attributes. Successive congruent models should be able to explain previous models (Hendry and Krolzig, Citation2001).