Methods and models of real estate mass appraisal

Abstract

This article examines the approaches, methods and models of mass appraisal. The article also considers features of application of valuation approaches and methods when conducting mass appraisal taking into account the type of real estate, the impact of price-forming factors of demand and supply and etc. Based on the analysis of scientific works of foreign and domestic scientists, the classification of mass appraisal models is given, which includes parametric multiple regression models, spatial and nonparametric models, as well as more modern methods, in particular, artificial neural networks, genetic algorithms, rough set theory and etc. Hedonistic models as the most common type of parametric multiple regression have been studied. In particular, the mathematical interpretation of this model is given, the key pricing factors that are used as explanatory variables in the construction of the model are analyzed, as well as different types of hedonistic models, their advantages and disadvantages are considered. As complementation of the traditional hedonistic model, spatial models are investigated , while nonparametric models are studied as an alternative. In particular, foreign scientists’ works are analyzed, which consider the comparative analysis of the effectiveness of use of the above-mentioned models. The article also considers modern methods, in particular artificial neural networks, genetic algorithms, rough set theory and expert models, the use of which in the context of mass appraisal is the subject of discussion among scientists. Based on the results of the analysis, criteria are formulated which impact the choice of application of specific mass appraisal methods and models. Further research will be directed towards a more detailed study of the hedonistic model in order to develop the latter as a basic model of mass appraisal on the example of the residential real estate market of one of the districts in Kyiv.