Forecasting British Tourist Arrivals to Balearic Islands Using Meteorological Variables and Artificial Neural Networks

There is a clear understanding of the benefits of getting accurate predictions that allow diminishing the uncertainty inherent to the tourism activity. Managers, entrepreneurs, politicians and many other agents related to the tourism sector need good forecasts to plan an efficient use of tourism-related resources. In spite of the consensus on this need, tourism forecasters must make an even greater effort to satisfy the industry requirements. In this paper, the possibility of improving the predictive ability of a tourism demand model with meteorological explanatory variables is investigated using the case study of monthly British tourism demand to the Balearic Islands (Spain). For this purpose, a transfer function model and a causal artificial neural network are fitted. Meanwhile, the results are compared with those obtained by non-causal methods: an ARIMA model and an autoregressive neural network. The results seem to indicate that adding meteorological variables can increase the predictive power but, however, the most accurate prediction is obtained using a non-causal model, specifically an autoregressive neural network.