2015 | OriginalPaper | Buchkapitel
Rating Systems: A Dual Track for Screening and Monitoring
verfasst von : Michele Modina
Erschienen in: Credit Rating and Bank-Firm Relationships
Verlag: Palgrave Macmillan UK
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
The information needed in order to estimate the value of default probability is elaborated in a different manner (statistical-quantitative approach vs. judgmental approach), noting, in this manner, the segment of clientele considered each time. For the evaluation of retail clientele, the use of automatic systems prevails: each piece of available information has a certain weight based on a statistical model. The significance of the qualitative information, in order to estimate the default, is gathered by associating it to a numerical value to insert in the statistical models. The judgement obtained can be subject to opportune modifications by expert analysts, if it is not considered suitable in representing the level of risk of the counterpart. This methodology is appropriate for segments of clientele with insignificant income margins. On the contrary, in segments of clientele considered more important in terms of the applicant’s dimensions, the reliability of the statistical techniques tends to decrease due to the different characteristics of the counterparts and, for this reason, it is necessary to accompany the statistical judgement with an interpretation by expert analysts in the credit sector.