Modelling Operational risk Severities with Kernel Density Estimation Using the Champernowne Transformation
Company: Royal & Sun Alloance
Year Of Publication: 2006
Month Of Publication: February
Pages: 47
Download Count: 6
View Count: 1863
Comment Num: 0
Language: English
Source: working paper
Who Can Read: Free
Date: 7-22-2010
Publisher: Administrator
Summary
The subject of this paper is a one-method-¯ts-all approach to quantify and predict future
losses in insurance. This method is based on a semiparametric estimator which is corrected
by some nonparametric smoothing techniques. A number of alternative kernel functions are
considered for removing boundary bias, resulting from transforming data with a parametric
function to bounded support. We also analyse the crucial point of bandwidth selection in
nonparametric statistics and discuss three di?erent bandwidth methods, where two are new to
the ¯eld. An extensive simulation study is presented between totally eighteen di?erent kernel
density estimators, and we also consider a practical application based on operational risk data.
Operational risk itself is de¯ned as the risk of loss arising from inadequate or failed internal
processes, people and systems or from external even
losses in insurance. This method is based on a semiparametric estimator which is corrected
by some nonparametric smoothing techniques. A number of alternative kernel functions are
considered for removing boundary bias, resulting from transforming data with a parametric
function to bounded support. We also analyse the crucial point of bandwidth selection in
nonparametric statistics and discuss three di?erent bandwidth methods, where two are new to
the ¯eld. An extensive simulation study is presented between totally eighteen di?erent kernel
density estimators, and we also consider a practical application based on operational risk data.
Operational risk itself is de¯ned as the risk of loss arising from inadequate or failed internal
processes, people and systems or from external even
Author(s)
This document's citation network:
Similar Documents: