Reference : Measuring operational risk in financial institutions
Scientific journals : Article
Business & economic sciences : Finance
Measuring operational risk in financial institutions
Plunus, Séverine mailto [Université de Liège - ULg > HEC-Ecole de gestion de l'ULg > HEC-Ecole de gestion de l'ULg >]
Hübner, Georges mailto [Université de Liège - ULg > HEC-Ecole de gestion de l'ULg : UER > Gestion financière >]
Peters, Jean-Philippe mailto [Deloitte Luxembourg > > > >]
Applied Financial Economics
Yes (verified by ORBi)
[en] operational risk ; Basel II ; CreditRisk+ ; modelling
[en] The scarcity of internal loss databases tends to hinder the use of the advanced approaches for operational risk measurement (Advanced Measurement Approaches (AMA)) in financial institutions. As there is a greater variety in credit risk modelling, this article explores the applicability of a modified version of CreditRisk+ to operational loss data. Our adapted model, OpRisk+, works out very satisfying Values-at-Risk (VaR) at 95% level as compared with estimates drawn from sophisticated AMA models. OpRisk+ proves to be especially worthy in the case of small samples, where more complex methods cannot be applied. OpRisk+ could therefore be used to fit the body of the distribution of operational losses up to the 95%-percentile, while Extreme Value Theory (EVT), external databases or scenario analysis should be used beyond this quantile.
Researchers ; Professionals ; Students

File(s) associated to this reference

Fulltext file(s):

Restricted access
OpRisk.pdfPublisher postprint547.21 kBRequest copy
Open access
2012_Plunus_Hubner_Peters_preprint.pdfAuthor preprint240.75 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.