Title: Multivariate VaRs for operational risk capital computation: a vine structure approach
Authors: Dominique Guégan; Bertrand K. Hassani
Addresses: Université Paris 1 Panthéon-Sorbonne, CES UMR 8174, 106 boulevard, de l'Hopital, 75647 Paris Cedex 13, France ' Santander UK plc, 2 Triton Square, London, NW1 3AN (TS 6A 01), UK; Université Paris 1 Panthéon-Sorbonne, CES UMR 8174, 106 boulevard de l'Hopital, 75647 Paris Cedex 13, France
Abstract: The Basel advanced measurement approach requires financial institutions to compute capital requirements on internal datasets. In this paper, we introduce a new methodology permitting capital requirements to take into account the embedded dependence structures of operational risks. The loss distributions are provided in a matrix of 56 cells. Constructing a vine architecture, which is a bivariate decomposition of a n-dimensional structure (n > 2), we use this approach to compute multivariate operational risk VaRs. We analyse the results and compare them with classical methodologies based on LDF modellings. Our method is simple to carry out, easy to interpret and complies with the new Basel Committee requirements.
Keywords: operational risks; vine copula; loss distribution function; LDF; nested structure; multivariate VaR; value at risk; capital requirements; financial institutions; modelling.
DOI: 10.1504/IJRAM.2013.057104
International Journal of Risk Assessment and Management, 2013 Vol.17 No.2, pp.148 - 170
Received: 26 Apr 2012
Accepted: 05 Jun 2013
Published online: 19 Jul 2014 *