The Role of Credit Risks in Forecasting a Bank's Liquidity

This study is devoted to studying the importance of effective credit risk management in forecasting bank liquidity. Correct assessment of credit risks is an important factor in ensuring the financial stability, long-term profitability of banks and the stability of the financial system as a whole. The ability of banks to distinguish customers exposed to credit risks by their quality level allows them to identify factors that negatively affect liquidity in advance. The study analyzes the possibilities of using machine learning algorithms in forecasting credit risks and considers their application in bank decisions. In particular, the effectiveness of the Random Forest and Gradient Boosting models in determining credit risks is evaluated. These models identify the main factors affecting bank liquidity and reveal their importance in the decision-making process. Also, the transparency of the model results is ensured using an interpretable artificial intelligence approach. The results of the study serve to develop scientific and practical recommendations for banks in managing credit risks and forecasting liquidity.