Keywords: tourism, seasonality, seasonal adjustment, forecasting, unit roots
The majority of tourism destinations appears to be characterized by a remarkable degree of seasonality of demand that causes several problems to local firms and administrations, as it hampers the efficient use of available facilities and the development of local capabilities and of the job market. The wide array of issues related to seasonality of demand for tourism destinations has thus stimulated economic research. This paper briefly discusses the essential aspects of seasonality of tourism services demand and describes the main characteristics of the monthly tourist nights in the Province of Lecce, between January 1988 and December 2008. In addition, the seasonality of tourism services in this area is analyzed from an econometric point of view. Two different approaches are followed. The first one consists in the estimation of a SARIMA model based on the original data. The second one is based on seasonal adjustment procedures (TRAMO-SEATS and X12-ARIMA) that eliminate the seasonal component from the original time series. The adjusted data are fitted with ARIMA models. Finally, the forecasting achieved from the estimated models are compared in order to understand the impact of seasonality on the ability to forecast the future pattern of tourism demand.