Document Search
Add To My Bookshelf Sign in or Register Save And Annotate
Keywords:

operational sign in to follow this
copula sign in to follow this
extreme sign in to follow this
value sign in to follow this
expected sign in to follow this
shortfall sign in to follow this
Pareto sign in to follow this
Categories:

VaR Uses sign in to follow this
--Operational Risk sign in to follow this
Half-Life:
Impact:
Discuss This Paper
Sign in to follow this page
Recent Comments
  more
A Copula-Extreme Value Theory Approach for Modelling Operational Risk
Year Of Publication: 2003
Month Of Publication: September
Pages: 18
Download Count: 3107
View Count: 14390
Comment Num: 0
Language: EN
Source:
Who Can Read: Free
Date: 9-30-2003
Publisher: Administrator
Summary
The object of this work is to present a model for measuring the operational risk of a bank. This model utilizes a Monte Carlo simulation in order to determine the lossdistribution and the relative risk measures like Value-at-Risk (VaR) or ExpectedShortfall. The methodology presented in this work is based on a frequency/severitymodel traditionally used in the actuarial sciences for modelling the loss distribution in damage insurances. In particular, the severity for each business line/risk type ismodelled through a distribution log-normal in the centre and in the left tail, while theright tail is built using the principles of Extreme Value Theory. In this way, we mayefficiently model the right tail of the distribution which more affects the risk measures,avoiding every possible underestimate of risk. Moreover, while most of the traditionalmodels assume a perfect dependence of the losses among each risk type, and calculatethe total VaR by simply summing the VaRs of each business line/risk type, weexplicitly consider a more realistic dependence structure using copula functions. In this way we may reduce the capital requirement. We implement the model to catastrophe insurance loss data of three different business lines and compare the resultant risk measures with the ones calculated assuming both a simple log-normal distribution for the loss severity and a perfect dependence among the losses of the three business lines.
Author(s)
Di Clemente, Annalisa Sign in to follow this author
Romano, Claudio Sign in to follow this author
This document's citation network:
Similar Documents:
Close window
Sign up in one step, no personal information required. Already a Member?



Email:
Repeat Email:
User Name:
Password:
Confirm Password:

Sign Up


Welcome to GloriaMundi!
Thanks for singning up



continue or edit your profile