Seminario 21/21: Gabriel Brida (Universidad de la República. Uruguay). Compositional data analysis to model tourist expenditure by categories

Información

This study models the relative expenditure of tourists in terms of budget allocation as a function of a set of co-variables. The purpose is to model how the tourist distributes its budget among the different items, what are the characteristics of the tourist that infer a certain relative spending pattern and if this pattern has any differential behavior depending on the seasons of the year. Incoming tourists to Uruguay are analyzed by means cross-sections of official statistics data (2014–2019), the compositional data analysis (CODA) and a Dirichlet regression model. The study confirms that expending pattern in accommodation, food and others varies according to the season, the tourist destination and the type of accommodation. We found significant variations in the proportion of spending on shopping or other expenses when tourism does not take place in high season or when the tourist destination is neither Montevideo nor in the coast. The study brings tools to analyze the behavior of tourists visiting Uruguay as consumers as related to expenditure pattern at microeconomic level.