Tid: 14 mars 2005 kl 1515-1600
Plats : Seminarierummet 3733, Institutionen för matematik, KTH, Lindstedts väg 25, plan 7. Karta!
Föredragshållare: Babak Soltani.
Titel: Estimating loss distribution for operational risk using internal and external databases. (Examensarbete)
Sammanfattning: Quantification of operational risk has received increasing attention under the new Basel proposal. The proposal provides some possibility for banks to use internal databases to estimate their operational risk and the associated capital needed for unexpected losses. It is however accepted that using only internal databases may not result in an accurate capital charge calculation and internal data should be supplemented by external data in order to have a data sample which gives a correct capital charge. Mixing internal data with external data in a proper way is the object of this thesis which begins with a short introduction of different approaches for computing loss distributions. Than a presentation of Credibility theory which provides the necessary mathematical tools for the mixing process is submitted. Extreme Value Theory will be presented briefly followed by testing a popular assumption that under some circumstances make the mixing process feasible. Finally some results and plots will be delivered.