1. Introduction
Over the last forty years, the world’s tourism industry has
become increasingly competitive as travel and tourism have
become more accessible to ever greater numbers of consumers.
This has increased the economic opportunities for developed and
emerging destinations. The UNWTO predicts that international
tourism demand will double by 2020 to 1.6 billion visitors generating
nearly US $2 trillion in economic activity (Levy & Hawkins,
2010). Tourism can make major contributions to nations’
economic development, trade performance and prosperity, so long
as destinations are successful in attracting and serving increased
tourism demand. Therefore, it is very important for the agencies
involved in tourism and related industries to monitor and anticipate
trends in international demand and to use this knowledge for
effective and resourceful decision-making and planning. Forecasts
of tourism volume are a major prerequisite that enable destinations
to predict infrastructure development needs (Sheldon, 1993).
The last couple of decades have seen many studies of international
tourism demand forecasting by academic researchers,
industry practitioners and government organisations. These studies
are dominated by quantitative approaches which are often subdivided
into causal econometric models and time series models.
The former attempt to establish relationships between variables
such as tourism demand (as measured by numbers of tourist