Comparison of Selected Models of Credit Risk

Abstract

Abstract Credit risk presents the probability of loss that the company incurs in the event of a business partner (the counterparty) default. The default may occur if the liabilities are not met under the terms of the contract which in turn results into the loss of the company (the creditor). Specifically, the liabilities arose from the credit, trade or investment activities, payment system and trade settlement. Difficulties in credit risk modelling arise due to the fact that the company default is not a frequent phenomenon but it occurs mainly unexpectedly. However, if the default occurs, it often causes the creditors major losses which size cannot be quantified in advance. The issue of modelling and quantification of credit risk is the subject of interest of many studies, scientific articles and publications. The access of individual authors to the present issue is diverse and so the methodology used for this purpose is not uniform. The present contribution will address the analysis and comparison of four basic approaches of description, but especially the quantification of credit risk: CreditRisk+, Credit Metrics, Merton model and Credit Grades. The comparison will be made on the basis of the computer performance, the applicability to different types of companies (public or non-public tradable), the volatility of credit events, thecorrelation of credit events occurrence, the required input data, currency of data and such like. Conclusions and recommendations for the application of the various approaches in specific situations will be parts of the contribution.