Reference : A Markov-switching Generalized additive model for compound Poisson processes, with ap...
Scientific journals : Article
Business & economic sciences : Quantitative methods in economics & management
http://hdl.handle.net/2268/216968
A Markov-switching Generalized additive model for compound Poisson processes, with applications to operational losses models
English
Hambuckers, julien mailto [Université de Liège - ULiège > HEC Liège : UER > Statistique appliquée à la gestion et à l'économie >]
Kneib, Thomas mailto []
Langrock, Roland mailto []
Silbersdorff, Alexander mailto []
2018
Quantitative Finance
Routledge
18
10
1679-1698
Yes (verified by ORBi)
International
1469-7688
1469-7696
[en] GAMLSS ; Hidden Markov model ; operational losses ; generalized Pareto distribution ; compound Poisson process ; distributional regression
[en] This paper is concerned with modeling the behavior of random sums over time. Such models are particularly useful to describe the dynamics of operational losses, and to correctly estimate tail-related risk indicators. However, time-varying dependence structures make it a difficult task. To tackle these issues, we formulate a new Markov-switching generalized additive compound process combining Poisson and generalized Pareto distributions. This flexible model takes into account two important features: on the one hand, we allow all parameters of the compound loss distribution to depend on economic covariates in a flexible way. On the other hand, we allow this dependence to vary over time, via a hidden state process. A simulation study indicates that, even in the case of a short time series, this model is easily and well estimated with a standard maximum likelihood procedure. Relying on this approach, we analyze a novel dataset of 819 losses resulting from frauds at the Italian bank UniCredit. We show that our model improves the estimation of the total loss distribution over time, compared to standard alternatives. In particular, this model provides estimations of the 99.9% quantile that are never exceeded by the historical total losses, a feature
particularly desirable for banking regulators.
Chair of Statistics and Econometrics, Faculty of Economics
Deutsche Forschungsgemeinschaft - DFG
RTG 1644 Scaling problem in Statistics
Researchers ; Professionals ; Students ; General public
http://hdl.handle.net/2268/216968

File(s) associated to this reference

Fulltext file(s):

FileCommentaryVersionSizeAccess
Open access
HMM_GAM for GPD regression.pdfSupplementary material available upon demandAuthor preprint654.35 kBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.