Abstract
Abstract This chapter examines the application of the credibility theory , originally developed in the field of risk theory and insurance mathematics, to benchmarking of credit ratings from heterogenous and different credit portfolios. This approach offers an interesting theoretical framework in which robust ratings comparability is enabled by explicitly correcting risk bucket estimates for the portfolio’s structure effects that are deemed significant. Moreover, statistical tests may then be performed to assess the comparability of whole different risk structures reflected in rating systems. Finally, hierarchical credibility models dealing with nested risk classifications and estimations may also be useful to formalize mapping rules in particular when the benchmark is not of the same granularity. Such ‘Credible’ mapping could then help build more consistent master scales.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
