Modelling demand for tourism in Italy

The aim of this thesis is to construct and estimate the demand for tourism for the Italian Province of Sassari, in the Sardinian island, and Italy as a whole. Several propositions are investigated. A systematic understanding is carried out for separating the domestic from the international demand of tourism to Sassari Province. The historic evolution of tourists' flows, the seasonality, trading-day effects and the empirical findings from the econometric investigation validate the separation of the two components. The sample period under estimation is from 1972 up to 1995. Three dynamic models are estimated at monthly, quarterly and annual frequencies for Sassari Province; analogies and differences are explored amongst the three models. On balance, the evidence indicates that the monthly and quarterly data models are superior to annual data models. However, one does not want to omit the annual estimation. Ideally, one should integrate and learn from each of the separate analysis. Some of the recently developed econometric techniques are used. A pre-modelling data analysis is undertaken for the economic series of interest. Seasonal and long run unit roots tests have given insight on the properties of the variables under study. The Johansen cointegration analysis is used in order to examine possible long run relationships amongst variables integrated of order one. Dynamic estimations are run in terms of the number of tourists for Sassari Province and monthly data expenditure for Italy. The LSE general-to-specific methodology is followed and a full range of diagnostic tests is provided. Short and long run income elasticities, negativity and substitutability are tested in the light of economic theory and other empirical studies existing in the tourism literature.