Forecasting the Demand for International Business Tourism

Abstract

Previous research in the area of tourism demand modeling and forecasting has paid little attention to business tourism. This study provides the most comprehensive comparison to date of the accuracy of modern forecasting methods in the context of international business tourism demand forecasting. Seven forecasting models are examined, including the error correction model and various structural time-series and autoregressive integrated moving average (ARIMA) models. The empirical results show that relative forecasting performance is highly dependent on the length of forecasting horizon, that adding explanatory variables to the structural time-series model does not improve forecasting performance, and that testing for unit roots is likely to yield reasonably accurate results under certain conditions.

Forecasting tourist arrivals using time-varying parameter structural time series models

“…Cycles do not feature in any markets over the observed sample period. This is consistent with most of the tourism demand literature using CSMs, such as the studies of González and Moral (1995) and Kulendran and Witt (2003). It indicates that the key economic determinants included have successfully explained the variation in the tourism demand variable, and therefore it is not necessary to include an additional cyclical component (Moosa, 2000).…”

“…Both the BSM and the CSM have been applied in the tourism demand forecasting context, and several studies have demonstrated their superior forecasting performance relative to other time series alternatives (see for example González & Moral, 1995, 1996Kulendran & King, 1997;Kulendran & Witt, 2001;Turner & Witt, 2001). However, in the studies of both Turner and Witt (2001) and Kulendran and Witt (2003), the CSM produced less accurate forecasts than the BSM.…”

“…The seasonal naïve "No Change 1" (constant value for each season) and "No Change 2" (constant growth rate for each season) models, the SARIMA model, BSM, CSM, the TVP model, and ADLM are included in the comparison of the forecasting performances. Due to space constraints, the specifications of these models are omitted here, but technical details are given by Frechtling (2001) for the seasonal naïve models and Kulendran and Witt (2003) and Song and Witt (2000) for the rest.…”

A Bound Test Approach to Cointegration of Tourism Demand

“…The consumer price indices were used as a proxy for the cost of tourism in Malaysia relative to the cost of living in Australia adjusted by the exchange rate [19][20][21] . We therefore expect a negative sign for this variable.…”

Modeling and Forecasting of International Tourism Demand in ASEAN Countries

“…In this study, Naïve method was used. SES model (Chen et al, 2008;Witt et al, 1994), SMA model (Makridakis et al, 1998;Hu et al, 2004;Lim and McAleer, 2008), Box-Jenkins model (Makridakis and Hibon, 1979), ARIMA (Kim et al, 2011;Goh and Law, 2002;Preez and Witt, 2003), Holt's DES model (Lim and McAleer, 2001;Makridakis et al, 1998), BSM model (Greenidge, 2001;Gonzalez and Moral, 1995;Turner and Witt, 2001;Kulendran and Witt, 2003) are other time series methods used by scholars. Casual econometric methods are other quantitative forecasting methods.…”