Models of multiple regression as an instrument of regional mortgage markets’ functioning effectiveness assessment

Tatyana S. Korosteleva,Vladimir E. Tselin

DOI: https://doi.org/10.18287/2542-0461-2021-12-3-164-178

2021-11-25

Abstract:The article is dedicated to issues related to the formation of the Russian Federation regions economically justified classifications by housing mortgage lending (HML) systems stage of development, for the implementation of mortgage markets selective government adjustment policy. The purpose of the study is to develop the methodology of regions mortgage potential assessment based on the search of additional application solutions in the area of rapid assessment and forecasting of regional mortgage markets state. To achieve the purpose, an additional version of common methodology of Russian Federation regions systematic ranking by HML systems stage of development is represented, and its main steps are structured. Within the third step of new methodology, a multiple regression model of Factor 1 (F1) is developed, which was interpreted in authors early studies as a performance indicator of regional mortgage markets functioning effectiveness. The models development was based on the statistical processings results of 510 observations in 85 Russian Federation regions in a period from 2014 to 2019. The high accuracy of models approximation has been proved, as well as its statistical reliability. The application capabilities of the model are represented, particularly the results of period 2020-2021 forecasting for Samara region and Saint-Petersburg city. The ways of managing the rating of region through targeted distribution of available regional and federal resources received by the models predictors are outlined. The novelty of this study is contained in the results of adaptation of the new assessment methodology, which is based on the obtainment of multiple regression model for Factor 1. The methodology base of the study consists of systematic analysis methods and multivariate statistics, particularly regressive and correlation analysis methods.