Christine A. Martin

Econometric Models for Forecasting International Tourism Demand

This article presents a set of econometric models for forecasting interna tional tourism demand. The models are developed from tourist visits from West Germany and the United Kingdom to their respective tourist destinations.

Substitute prices in models of tourism demand

Abstract Models of international tourism demand which incorporate substitute prices as explanatory variables are specified. In tourism there are two price components, the transport cost to the destination and the living cost in the destination, and hence it is necessary to allow for both costs to and in substitute destinations. The substitute price variables are constructed using a weighting system based on market shares of major competing destinations, but the weights are allowed to alter throughout the estimation period to cater for changing trends. The empirical results support the hypothesis that substitute prices play an important role in determining the demand for international tourism, but there is considerable variation in importance according to the origin under consideration and the mode of transport

Forecasting tourism demand: A comparison of the accuracy of several quantitative methods

Abstract Forecasting accuracy is examined in the context of international tourism demand. Seven quantitative forecasting methods are used to generate out of sample forecasts for tourist flows across twenty-four origin-destination pairs, and two forecasting horizons. Two alternative accuracy measures are used to evaluate forecasting performance. Statistically significant differences in forecasting accuracy are identified using the ANOVA and Scheffe tests. Several of the simple forecasting methods produce more accurate forecasts than econometric forecasts. As expected, one-year-ahead forecasts are more accurate than two-years-ahead forecasts. Aggregation of data series appears to reduce forecasting accuracy slightly.

Accuracy of econometric forecasts of tourism

Abstract Many papers present estimated econometric models of international tourism demand. Their authors often suggest that these may be used for short-term forecasting, but actual forecasts are rarely provided. This paper presents the results of a study in which the forecasting accuracy of several quantitative techniques is compared. Statistically significant differences in accuracy are identified using the ANOVA and Scheffe tests. Several of the simple methods produce more accurate forecasts than econometrics; this finding is supported by an examination of the accuracy of commercially available econometric forecasts of tourism, where the naive “no change” model forecasts more accurately in 70% of cases.

International tourism-demand models - inclusion of marketing variables.

Abstract National tourist organizations often spend considerable sums in foreign countries on promoting the particular country as a tourist destination. Only seldom, however, is marketing activity incorporated as an explanatory variable in models of the demand for international tourism. In such cases, it should be possible to assess the impact of promotional expenditure in terms of the tourist visits and tourist expenditure generated. Where marketing is used to explain international tourism demand, caution must be exercised in interpreting the empirical results; poor results cannot be used to reach sensible conclusions, and when good results are obtained the full implications of the estimated coefficients need to be explored.

Deriving a Relative Price Index for Inclusion in International Tourism Demand Estimation Models: Comment:

The demand for international tourism depends on many external factors such as international economic prospects, international political climate, air accessibility, foreign hotel investment and management, or seasonal oscillation, to name a few. International and even regional change in any of these factors, which could have major influence on the foreign tourists’ arrivals, have limited influence on domestic tourism.