USING CLUSTER ANALYSIS TO ASSESS FINANCIAL STABILITY AS AN OBJECT OF MANAGERIAL IMPACT OF REGIONAL COMPETITIVE IMMUNITY

The relevance of considering and analyzing financial stability and competitive immunity at the meso-level in modern conditions is increasing due to changes in the state of both the economic and social spheres. It was found that the “competitive immunity of the territory” reflects a number of new characteristics of modern territorial-regional-interregional competition in the global economy, which distinguishes it from the concept of economic security both at the macro and meso levels. The paper considers the category of “competitive immunity of the region”, which implies the possibility of survival of the peripheral territories of the regions of Ukraine and maintaining their high level of competitiveness. In accordance with the accepted concept of competitive immunity, three problemarea blocks were identified: information-digital approach; information and digital technologies; cost and reputation management, which include objects of managerial influence necessary to evaluate the transition of competitive immunity to sustainable functioning. The main aspect in the study of the financial stability of the local regional budgets as an integral part of the competitive immunity of the region was the search for criteria and the development of a methodology for evaluating efficiency. The following performance indicators of local budgets were used: budget revenues; budget spending; intergovernmental transfers from the state budget; tax revenues; the amount of equalization subsidies; non-tax revenues; average population. An applied study of the methodology for assessing the financial sustainability of the budget as an object of managerial influence at the local level was carried out on the example of selected indicators of local budgets of all regions of Ukraine for 2018-2020. The calculation of the selected indicators was made on the basis of statistical data on the local budgets implementation, reports and decisions of regional councils on the regional budget. The distribution of the initial data set into clusters was analyzed with help of the Deductor business analytical platform, using the k-means clustering algorithm and Kohonen maps. Based on the results of the k-means algorithm, it was found that it is advisable to divide the sample for classifying regions into three groups. To compare and evaluate the effectiveness of the results obtained, as well as to supplement the analysis of the financial stability of the regions of Ukraine, Kohonen maps were used using the Deductor business analytical platform. It was revealed that both methods allow efficient clustering of data in a multidimensional space. The results of clustering obtained by different methods are consistent with each other and, when applied in a complex manner, make it possible to classify the elements of the sample with maximum likelihood and minimum error. The regions of Ukraine were grouped according to the financial stability of the local budget into three groups: regions with high financial stability, regions with medium financial stability and regions with low financial stability. The correct interpretation of the results obtained through a comprehensive analysis of financial stability in relation to the local budget using clustering or using neural networks allows not only to analyze the obtained values, but to compare them with the standard and conduct a comparative analysis relative to other regions, identify the influence of factors on the change in the integral indicator, give a predictive assessment for the future and justify the chosen strategy for strengthening competitive immunity for a particular region.