Purpose
The purpose of this paper is to develop an innovative model that can be included within the market approach methods for property valuations. The algorithm takes into account the frequent high level of dissimilarity of the comparables selected for the assessment, thus providing for the use of appropriate similarity and reliability coefficients capable of weighing the data of the comparison sample with respect to the different degrees of similarity and reliability.
Design/methodology/approach
The proposed model borrows the operative logics of the goal programming techniques, in order to identify the solution, the market value of the subject property and the implicit prices of the different influencing factors, since they are more reliable from the mathematical and empirical points of view.
Findings
The model has been applied to two case studies, relating to samples of residential properties located in the city of Naples (Southern Italy). The results obtained have outlined the high valuation performance of the developed appraisal model, capable of overcoming the applicability limits of classical market approach methods as well as providing solutions that are highly consistent with the expected empirical phenomena.
Practical implications
The research takes into account the growing need of both professionals and end users (banks, courts, public and private Entities, etc.) for valuation models that are easily repeatable and sufficiently objective. They are required in order to allow for the rapid verification of the elaborations carried out as well as to check the valuer’s appreciation of the contribution of the influencing factors in the market price formation. The outputs of the two applications developed have highlighted the ability of the proposed model to satisfy these market requests.
Originality/value
The proposed model can be easily implemented through a simple calculation program, with the mathematical structure elaborated allowing to overcome some application limits of the classical market approach methods. Furthermore, the introduction in the algorithm of appropriate similarity and reliability coefficients, capable of suitably weighting the data of the comparison sample, allows to widen the spatial horizon for the identification of the comparables as well as select properties characterized by a high level of dissimilarity. This makes it possible to apply the model in territorial contexts characterized by markets that are not excessively dynamic.