The Use of Credit Scoring in Credit Risk Assessment

This bachelor's thesis deals with the use of credit scoring in credit risk assessment. Credit risk represents one of the most significant risks faced by financial institutions, and its accurate measurement is crucial for effective lending decisions. The thesis describes the nature of credit risk, its measurement, and prediction, with particular attention given to scoring models. It focuses especially on logistic regression, which is a commonly used tool for estimating the probability of borrower default. In the empirical part, a scoring model is applied to a dataset containing information about loans and clients. After data preprocessing, a logistic regression model is estimated and its predictive performance evaluated. The results show that although the model achieves only limited predictive accuracy, it allows identification of variables that significantly influence the probability of default. The thesis also discusses the limitations of the chosen approach and suggests possibilities for further development of scoring methods, including handling imbalanced data and a broader range of predictors.