COMPARATIVE ANALYSIS OF SUBSIZED REGIONS OF THE RUSSIAN FEDERATION USING MACHINE LEARNING METHODS FOR A WIDE RANGE OF INDICATORS OF FIXED ASSETS

The paper presents an original method for finding a connection between the subsidization of the regions of the Russian Federation and the factors of fixed assets provided by Rosstat reports. Within the framework of the proposed method, an analysis of indicators for 2021 was carried out. A set of the most significant indicators of fixed assets was identified, according to which the group of subsidized regions of the Russian Federation differs from the subjects of the Federation without subsidies. The following differences were revealed by machine learning methods in the compared groups: lower values in subsidized regions for such indicators as the cost of fixed assets at the end of the year at full book value (million rubles), water supply, wastewater disposal, organization of waste collection and disposal, pollution elimination activities, water supply; wastewater disposal, organization of waste collection and disposal, commissioning of enterprises with pollution elimination activities, manufacturing, commissioning of fixed assets by type of economic activity, wholesale and retail trade. At the same time, higher values in subsidized regions in terms of activity structure are observed in the field of information and communications, enterprises in the field of agriculture, forestry, hunting, fishing and fish farming. This approach has not been used by anyone in data analysis before. The found patterns will allow us to develop an action plan for the withdrawal of regions from the subsidy zone.

machine learning methods, statistics, subsidized regions, fixed assets, correlation coefficient, cluster analysis, mathematical modeling, features, partitions, subsidies

Kuznetsova A. V., Borisova L. R., Kremer N. S., Fridman M. N. Comparative analysis of subsized regions of the Russian Federation using machine learning methods for a wide range of indicators of fixed assets. Biznes. Obrazovanie. Pravo = Business. Education. Law. 2025;1(70):20—28. DOI: 10.25683/VOLBI.2025.70.1182.