This paper introduces a new model that takes inspiration from the approach of Davis and Lo (2001) but differentiates from it by relaxing the homogeneous assumption and by introducing few restrictions on the shape of the infection mechanism that allow for a good level of tractability. In particular, the contagion mechanism proposed is the result of two independent components: an infection attempt generated by defaulting frms and a failed defense from healthy ones. Within this framework it is possible to prove several theoretical results regarding marginal default probabilities and joint default/survival events. Finally, we provide an effcient recursive algorithm for the portfolio loss distribution similar, in spirit, to the one commonly used for CID (conditionally independent) models.
FARINA, Gianluca, GIACOMETTI, Rosella, (2015). Portfolio loss modeling: an infectious framework 1(2016) - MEQ Quantitative Methods). Bergamo: Retrieved from http://hdl.handle.net/10446/53315
Portfolio loss modeling: an infectious framework
GIACOMETTI, Rosella
2015-12-01
Abstract
This paper introduces a new model that takes inspiration from the approach of Davis and Lo (2001) but differentiates from it by relaxing the homogeneous assumption and by introducing few restrictions on the shape of the infection mechanism that allow for a good level of tractability. In particular, the contagion mechanism proposed is the result of two independent components: an infection attempt generated by defaulting frms and a failed defense from healthy ones. Within this framework it is possible to prove several theoretical results regarding marginal default probabilities and joint default/survival events. Finally, we provide an effcient recursive algorithm for the portfolio loss distribution similar, in spirit, to the one commonly used for CID (conditionally independent) models.File | Dimensione del file | Formato | |
---|---|---|---|
Binder1_2016_qm.pdf
accesso aperto
Versione:
publisher's version - versione editoriale
Licenza:
Licenza default Aisberg
Dimensione del file
387.8 kB
Formato
Adobe PDF
|
387.8 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo