The Correlation Problem in Operational Risk

Frachot, Antoine and Roncalli, Thierry and Salomon, Eric (2004): The Correlation Problem in Operational Risk. Published in: Operational Risk (May 2004)

Abstract

This paper demonstrates that aggregate losses are necessarily low as long as we remain under the standard assumptions of LDA models. Moreover empirical findings show that the correlation between two aggregate losses is typically below 5%, which opens a wide scope for large diversification effects, much larger than those the Basel Committee seems to have in mind. In other words, summing up capital charges is in substantial contradiction with the type of correlation consistent with the standard LDA model.

• Item Type: MPRA Paper
• Language: English
• Keywords: operational risk; LDA model; severity correlation; frequency correlation; aggregate loss correlation
• Subjects: C - Mathematical and Quantitative Methods > C0 - General > C01 - Econometrics

References

Principles for the Home-host Recognition of AMA Operational Risk Capital, Basel Committee on Banking Supervision, January 2004

Evans, M., N. Hastings and B. Peacock [1993], Statistical Distributions, John Wiley & Sons

Feller, W. [1971], An Introduction to Probability Theory and Its Applications, Volume II, second edition, John Wiley & Sons

Frachot, A., P. Georges and T. Roncalli [2001], Loss distribution approach for operational risk, Working Paper, Crédit Lyonnais, Groupe de Recherche Opérationnelle

Frachot, A., O. Moudoulaud and T. Roncalli [2003], Loss distribution approach in practice, in M. Ong (ed.), The Basel Handbook: A Guide for Financial Practitioners, Risk Books

Johnson, N.L., S. Kotz and N. Balakrishnan [1997], Discrete Multivariate Distributions, Wiley Series in Probability and Statistics, John Wiley & Sons

Klugman, S.A., H.H. Panjer and G.E. Willmot [1998], Loss Models: From Data to Decisions, Wiley Series in Probability and Mathematical Statistics, John Wiley \& Sons