Applying of machine learning methods to evaluate the financial decentralization reforms in the context of ensuring inclusive and sustainable growth of regions

Abstract

Abstract The structure of the model basis for assessing the effectiveness of financial decentralization reforms, developed on the basis of such Artificial Intelligence Methods as Kohonen neural networks, methods of cluster analysis, taxonomy, principal components, panel data, production-institutional functions, convergence analysis methods is proposed. An assessment of the impact of financial decentralization reforms on the pace of socio-economic development of territories, on the convergence-divergence of regional development and on the resistance of regions to “shocks” was carried out. Models have been developed to assess the impact of the level of financial decentralization on the rate of convergence of regional development and the human development index using panel data analysis methods. Models of production and institutional functions have been developed to determine the potential for growth in resource efficiency by increasing the level of financial decentralization. Models for classifying the internal regions of Ukraine according to the level of financial decentralization and resource efficiency of production and economic systems have been developed. The proposed model basis can be used to build dynamic maps for assessing the effectiveness of financial decentralization reforms in order to monitor and adapt regional development strategies.