DEVELOPMENT OF PREDICTIVE MODELS OF LAND VALUE OF SINGLE-INDUSTRY TOWNS WITH ALLOWANCE FOR THE FACTORS OF ACTIVITY OF CITY-FORMING ENTERPRISES (ON THE EXAMPLE OF THE MURMANSK REGION)

Abstract

The real estate market and features of real estate price formation depend on a large number of economical, social, political, natural, demographic and other factors. In each separately taken settlement, the degree of influence of each group of factors is different, which is largely determined by the features of the city. The object of the present research is the land plots of Murmansk region monotowns with their city-forming enterprises whose economic activity has a significant effect on the price formation of the real estate, its market and cadastral value. The specified influence allows the price of the real estate to be predicted, depending on a change in indicators of the city-forming branch. The purpose of the paper is to construct and justify predictive models reflecting the dependence of Murmansk region monotowns land prices on economic factors of the city-forming branch. In evaluating the relationship between the assessment of land plots and the selection of significant indicators, the advantage of the neural network model is proved, as it has better predictive properties than regression models.