Territorial Determinants of Tourist Arrivals: A Multivariate Regression Tree Analysis

We investigate the territorial and structural determinants of tourism arrivals. Understanding which elements influence the number of tourists is crucial for correctly planning tourism development in a destination. The analysis is carried out using a Multivariate Regression Tree, which allows the identification not only of the main determinants of tourist arrivals but also of similarities among different geographic areas. The study focuses on 307 Sardinian municipalities. Tourist arrivals disaggregated by nationality are used as response variables, while geographic morphology, demographic structure, and transport accessibility are considered as explanatory variables. The results show that the density of accommodation facilities represents the primary determinant of tourism attractiveness. Moreover, they highlight a persistent coastal–inland dichotomy, suggesting that tourism concentration is strongly driven by the interaction between accommodation supply and accessibility. Overall, the study provides a nuanced understanding of how local territorial characteristics shape tourism demand across different international market segments, offering relevant insights for regional destination management.