Forecasting US overseas travelling with univariate and multivariate models
Name:
Forecasting US travelling.docx
Embargo:
2023-01-06
Size:
78.34Kb
Format:
Microsoft Word 2007
Authors
Apergis, NicholasAffiliation
University of DerbyIssue Date
2021-01-06
Metadata
Show full item recordAbstract
This study makes use of specific econometric modelling methodologies to forecast US outbound travelling flows to certain destinations: Europe, Caribbean, Asia, Central America, South America, Middle East, Oceania, and Africa, spanning the period 2000-2019 on a monthly basis. Both univariate (jointly with business conditions) and multivariate models are employed, while out-of-sample forecasts are generated and the results are compared based on popular forecasting performance criteria. These criteria show that in the case of univariate models, the largest forecasting gains are obtained when the modelling process follows the KS-AR(1) model with the business cycles being measured as the coincident indicator. In the case of multivariate models, the largest forecasting gains occur with the standard VAR model for very short forecasting horizons, and with the Bayesian VAR for longer horizons. The results are robust to both total and individual destinations. The findings allow interested stakeholders to gain insights into near-future US outbound tourism to popular diversified international destinations, as well as to better understand its positive and negative impacts for strategic planning and destination adaptation purposes.Citation
Apergis, N. (2020). ' Forecasting US overseas travelling with univariate and multivariate models'. Journal of Forecasting, pp. 1-29.Publisher
WileyJournal
Journal of ForecastingDOI
10.1002/for.2760Additional Links
https://onlinelibrary.wiley.com/doi/epdf/10.1002/for.2760Type
ArticleLanguage
enEISSN
1099-131Xae974a485f413a2113503eed53cd6c53
10.1002/for.2760
Scopus Count
Collections
The following license files are associated with this item:
- Creative Commons
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivatives 4.0 International