A large number of studies have been devoted to analyzing international tourism demand; however, even today, the impact of climate and weather on tourism receives only limited attention. We studied the empirical influence of the North Atlantic Oscillation (NAO), the most important mode of variability in northern atmospheric circulation, on international tourism demand -specifically, tourist arrivals from the UK and Germany to the Balearic archipelago (Spain). We used 2 traditional techniques usually applied in natural sciences, cross-correlation functions and the Granger causality test, as well as a novel and flexible methodology called the autoregressive distributed lag (ARDL) boundstesting approach. ARDL modeling can be a useful tool in illuminating relationships between variables. Our empirical evidence demonstrates the existence of a statistical relationship between the NAO and tourist arrivals from the UK and Germany to the Balearic Islands. The finding of a statistical relationship between the NAO and tourism suggests that this atmospheric phenomenon can be of great interest to social researchers who study international tourism flows. The NAO index can be used as an explanatory variable in tourism demand models, allowing researchers to develop parsimonious models, as well as to avoid certain economic problems (e.g. multicolinearity).
KEY WORDS: North Atlantic Oscillation · International tourism demand · Granger causality test · Cross-correlation analysis · Autoregressive distributed lag · ARDL · Balearic IslandsResale or republication not permitted without written consent of the publisher 1 Climate and weather are meteorologically similar phenomena manifested and studied at different scales. De Freitas (2003) defines climate as the accumulation of daily and seasonal weather events over a long time period. On the other hand, weather is the condition of the atmosphere at a specific place and at a specific time Clim Res 43: 207-214, 2010 2007, Scott et al. 2008). The most widely used variables were temperature, rainfall, wet days, cloud cover, humidity, sunshine, and wind speed. However, the inclusion of these variables in a tourism-demand model could lead to some econometric problems since they are highly correlated to each other (multicollinearity problems). Other researchers have constructed indices that allow summarizing the significance of climate for tourism, but they were only used to evaluate and rate recreational climates in terms of user sensitivity and satisfaction (de Freitas 2003.One key variable that might be useful to summarize meteorological information and potentially useful to explain tourism is the recurrent pattern of atmospheric variability observed over the North Atlantic Ocean, known as the North Atlantic Oscillation (NAO). The phenomenon is formally defined as an anomalous difference in atmospheric pressure between a subtropical high-pressure belt (around the latitudes of 35 to 40°in the Northern Hemisphere and centered near the Azores) and a subpolar low-pressure belt (cen...