An Application to a Spanish Case Study of a Property Valuation Models

In 2017 the European AVMAlliance emphasized the importance of Automated Valuation Methods, to be used to assess market values and/or to monitor the evolution of property prices. This last aspect has acquired a particular cogency for market operators (sellers, buyers, investors, etc.) over time, in order, on the one hand, to make reliable valuations and, on the other hand, to effectively and quickly check the trend of the values. The present research aims at analyzing the functional relationships between the unit selling prices and the explanatory variables that contribute to their formation. The study has been carried out on a sample of two-hundred and ten residential properties, sold in the period 2016–2017 and located in the city of Tarragona (Spain). The main factors considered by sellers and buyers in the preliminary negotiation phase have been collected. The implementation of a data-driven technique has allowed to identify a statistically reliable model which, in addition to highlight the most influencing factors, has shown the interdependences between the variables considered and the unit selling prices.