METHODOLOGICAL APPROACHES TO RISKS ASSESSMENT IN THE CREDIT SECURITY SYSTEM

Abstract

The article is dedicated to current issues of credit risk assessment and lending security in Ukraine under conditions of martial law, persistent macroeconomic instability, and heightened financial market volatility. The study systematically examines the dynamics of non-performing loans (NPLs) in bank credit portfolios over the period 2020–2025, analyzing their correlation with key macroeconomic indicators, including GDP growth rate, inflation, the National Bank of Ukraine's key policy rate, unemployment rate, and the hryvnia-to-US dollar exchange rate. A detailed correlation matrix was constructed to identify the strength and direction of relationships between NPL levels and the selected macroeconomic variables. Subsequently, a multivariate regression model was built to quantitatively assess the contribution of each factor to NPL dynamics. The results demonstrate that unemployment and inflation exert the strongest positive effect on NPL growth, with correlation coefficients of 0.954 and 0.787, respectively. GDP growth showed a weak negative correlation with NPL levels, suggesting that short-term economic fluctuations have a limited direct effect on portfolio quality. Exchange rate volatility displayed a moderate negative relationship, reflecting the partial influence of currency depreciation on foreign-currency-denominated loans. The NBU key policy rate showed a moderate positive correlation, indicating that tighter monetary conditions increase debt-servicing burdens for borrowers. The article systematically reviews modern methodological approaches to credit risk management, encompassing scoring models, stress testing, scenario analysis, and Big Data analytical tools. Practical implications for Ukrainian banks operating under wartime challenges and economic uncertainty are discussed in detail. The study emphasizes the importance of building an integrated risk management framework that combines advanced quantitative analytics, qualitative expert assessments, strict regulatory compliance, and robust internal controls. The findings provide a methodological basis for constructing predictive NPL models and optimizing bank credit portfolio strategies.