Forecasting U.K. Inbound Expenditure by Different Purposes of Visit

This source preferred by Shuang Cang

Authors: Cang, S. and Hemmington, N.

Journal: Journal of Hospitality & Tourism Research

Volume: 34

Pages: 294-309

ISSN: 1096-3480

DOI: 10.1177/1096348009350616

Accurately forecasting U.K. inbound expenditure by purpose of visit plays an important role in tourism planning and policy making. Forecasting U.K. inbound expenditure at the disaggregated level is studied in this article. Disaggregating is done on the basis of purpose of visit: holiday, business, study, visit friends or relatives (VFR), and miscellaneous.

The most robust two time series forecasting models, seasonal autoregressive integrated moving average (ARIMA) and Winters's multiplicative exponential smoothing (WMES), are applied in this article. The Naïve 2 forecasting model is used as a benchmark to compare with the ARIMA and WMES models. The outcomes of the forecasting results show that the ARIMA model outperforms the WMES model, but it is not statistically superior to the WMES model. The ARIMA and WMES models are both statistically superior to the Naïve 2 model for this U.K. inbound expenditure data set. The ARIMA model forecasts a higher increasing trend for expenditure than the WMES model for the business purpose, whereas the WMES model forecasts a higher increasing trend for expenditure than the ARIMA model for miscellaneous purpose. It is recommended that combining the values from the ARIME and the WMES models is used as forecasting values on these business and miscellaneous purposes.

This data was imported from Scopus:

Authors: Cang, S. and Hemmington, N.

Journal: Journal of Hospitality and Tourism Research

Volume: 34

Issue: 3

Pages: 294-309

eISSN: 1557-7554

ISSN: 1096-3480

DOI: 10.1177/1096348009350616

Accurately forecasting U.K. inbound expenditure by purpose of visit plays an important role in tourism planning and policy making. Forecasting U.K. inbound expenditure at the disaggregated level is studied in this article. Disaggregating is done on the basis of purpose of visit: holiday, business, study, visit friends or relatives (VFR), and miscellaneous. The most robust two time series forecasting models, seasonal autoregressive integrated moving average (ARIMA) and Winters's multiplicative exponential smoothing (WMES), are applied in this article. The Naïve 2 forecasting model is used as a benchmark to compare with the ARIMA and WMES models. The outcomes of the forecasting results show that the ARIMA model outperforms the WMES model, but it is not statistically superior to the WMES model. The ARIMA and WMES models are both statistically superior to the Naïve 2 model for this U.K. inbound expenditure data set. The ARIMA model forecasts a higher increasing trend for expenditure than the WMES model for the business purpose, whereas the WMES model forecasts a higher increasing trend for expenditure than the ARIMA model for miscellaneous purpose. It is recommended that combining the values from the ARIME and the WMES models is used as forecasting values on these business and miscellaneous purposes. © 2010 International Council on Hotel, Restaurant and Institutional Education.

The data on this page was last updated at 04:46 on November 24, 2017.