Measuring operational risk using a mean scaled individual risk model

摘要

Building on a new theory of parametric risk models initiated in Hürlimann [Blätter der Deutschen Gesellschaft für Versicherungsmathematik XXIII(3) (1998) 251], it is shown how mean scaled individual risk models can be constructed. The approximate computation of their distributions and related quantities can be done in the author's (1990) mathematical framework of generalized pseudo compound Poisson distributions, which are characterized by integral equations of Volterra type. The results by Dhaene and De Pril [Insurance: Mathematics and Economics 14 (1994) 181] are extended to this more general class of models. The method is applied to the construction of a mean scaled operational risk model.