WAYS TO IMPROVE THE MECHANISM FOR PREVENTING BANK INSOLVENCY: AN ECONOMIC MODEL FOR ASSESSING INSOLVENCY FACTORS

Abstract

In the current conditions of economic instability, military challenges and global financial shocks, the issue of bank insolvency is becoming particularly relevant. The stability of the banking system is a fundamental prerequisite for the effective functioning of the national economy, and timely identification and prediction of signs of financial instability of banks allows not only to minimize risks for depositors, but also to maintain confidence in the financial sector as a whole. The paper analyzes existing approaches to assessing the solvency of banks, in particular based on the CAMELS system, and also justifies the need to improve the methods through the integration of econometric modeling and machine learning methods. It is concluded that CAMELS, although it remains a standard for supervision, has a number of shortcomings – in particular, limited coverage of non-financial risks, inefficiency and subjectivity of assessment. In this context, it is proposed to build a comprehensive economic model for assessing bank insolvency factors using logistic regression and machine learning algorithms, such as decision tree, random forest, XGBoost and neural networks. The model takes into account financial, macroeconomic and behavioral indicators, which allows identifying problem banks at an early stage. A comparative analysis of the methods was carried out, their advantages and limitations were identified. Based on the analysis of examples of banks that became insolvent in Ukraine, the effectiveness of the multidimensional approach was confirmed. Recommendations were made for integrating the model into the monitoring system of the National Bank of Ukraine. The proposed model can become an effective tool for state regulation of the financial sector and increasing the country's financial security.