Analysis of tourism demand serial dependence structure for forecasting
Date of Issue2017
College of Business (Nanyang Business School)
This study aims to extend knowledge of serial dependence structure in tourism demand modelling and make a contribution to tourism forecasting with the use of copula method. Analysis of serial dependence can reveal the impact of current tourism demand on the future. This is important for tourism demand forecasting, as the prediction of future tourism demand relies highly on the historical demand information. However, serial dependence, especially its structure, has received very little attention in previous tourism research. The copula method is flexible as it provides various functions to specify different serial dependence structures and allows arbitrary distributions of tourism demand. We used five types of copulas to analyse two-dimensional serial dependence structure for 10 arrivals series to Singapore. The empirical findings show that serial dependence structures of arrivals can be non-linear. Additionally, the Student-t copula generates forecasts of tourism demand with higher accuracy than the autoregressive integrated moving average (ARIMA) and seasonal ARIMA models.
© 2017 The Author(s). This paper was published in Tourism Economics and is made available as an electronic reprint (preprint) with permission of IP Publishing Ltd. The published version is available at: [http://dx.doi.org/10.1177/1354816617693964]. One print or electronic copy may be made for personal use only. Systematic or multiple reproduction, distribution to multiple locations via electronic or other means, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper is prohibited and is subject to penalties under law.