Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/92775 
Year of Publication: 
2001
Series/Report no.: 
ISER Discussion Paper No. 533
Publisher: 
Osaka University, Institute of Social and Economic Research (ISER), Osaka
Abstract: 
This paper examines stationary and nonstationary time series by formally testing for the presence of unit roots and seasonal unit roots prior to estimation, model selection and forecasting. Various Box-Jenkins Autoregressive Integrated Moving Average (ARIMA) models are estimated over the period 1975(1)-1989(4) for tourist arrivals to Australia from Hong Kong, Malaysia and Singapore. The mean absolute percentage error (MAPE) and root mean squared error (RMSE) are used as measures of forecast accuracy. As the best fitting ARIMA model is found to have the lowest RMSE, it is used to obtain post-sample forecasts. Tourist arrivals data for 1990(1) to 1996(4) are compared with the forecast performance of the ARIMA model for each origin market. The fitted ARIMA model forecasts tourist arrivals from Singapore between 1990(1)-1996(4) very well. Although the ARIMA model outperforms the seasonal ARIMA models for Hong Kong and Malaysia, the forecast of tourist arrivals is not as accurate as in the case of Singapore.
Subjects: 
unit roots
seasonality
forecasting models
forecast accuracy
root mean square error
ex post forecasts
Document Type: 
Working Paper

Files in This Item:
File
Size
201.18 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.