Statistical methods in credit risk management

Successful banks base their operations on the principles of liquidity, profitability and safety. Therefore, the correct assessment of the ability of a loan applicant to carry out certain obligations is of crucial importance for the functioning of a bank. In the past few decades several credit scoring models have been developed to provide support to credit analysts in the assessment of a loan applicant. This paper presents three statistical methods that are used for this purpose in the area of credit risk management: logistical regression, discriminatory analysis and survival analysis. Their implementation in the banking sector was motivated to a great extent by the development and application of information and communication technologies. This paper aims to point out the most important theoretical aspects of these methods, but also to actualise the need for the development and application of the credit scoring model in Croatian banking practice