THE PROBLEMS OF INTERNAL AND EXTERNAL DATA USE FOR OPERATIONAL RISK ASSESSMENT IN COMMERCIAL BANKS
https://doi.org/10.32686/1812-5220-2016-13-4-72-79
Abstract
In this paper, authors consider several particular problems regarding data samples for operational risk assessment models calibration. This study analyzes some aspects of data partition into homogeneous groups. It holds qualitative analysis of grouping structure influence on resulting estimates. Further, it formulates three variants of the different threshold data mixing problem and then the algorithm of solving that data mixing problem is being proposed. As a result, authors raise the question of sufficiency of operational risk data for the model calibration. They consider three approaches to extrapolation beyond the data sample regarding the presence of extremal severity events. The conclusion of the paper points at the lack of operational loss data. In addition, authors suggest several ways of solving this problem.
About the Authors
L. V. Kokh
Peter the Great Saint-Petersburg Polytechnic University
Russian Federation
Russian Federation
S. M. Bulatsky
Peter the Great Saint-Petersburg Polytechnic University
Russian Federation
Russian Federation
References
1. Золотарева Е.Л. Математическое моделирование операционного риска в коммерческом банке: дис.. канд. экон. наук. Москва, 2011.
2. О порядке расчета размера операционного риска: положение Банка России от 03.11.2009 № 346-П (ред. от 03.07.2012) // Вестник Банка России. 2009. № 77.
3. Anghelache, G.-V. Operational Risk - An Assessment at International Level / G.-V. Anghelache, B.-O. Cozmanca, C.-A. Handoreanu, C. Obreja, A.-C. Olteanu, A.-N. Radu // International journal of mathematical models and methods in applied sciences. 2011. Vol. 1(5). C. 184-192.
4. Baud N., Frachot A., Roncalli T. How to avoid overestimating capital charge for operational risk? (February 2003) // Thierry Roncalli’s Home Page. URL: http://www. thierryroncalli.com/download/oprisktech.pdf (дата обращения: 10.02.2016)
5. Challenges in Measuring Operational Risk from Loss Data [Электронный ресурс]: офиц. сайт ORX. Цюрих, 2016. URL: https://www.orx.org/Lists/PublicDocuments/ Challenges_in_Measuring_OpRisk_from_Loss_ Data_9September2009.pdf (дата обращения: 15.12.2015).
6. Chernobai A., Rachev S., Fabozzi F. A Guide to Basel II Capital Requirements, Models, and Analysis. Hoboken: John Wiley & Sons, 2007. 300 p.
7. Frachot A., Roncalli T. Mixing internal and external data for managing operational risk (January 29, 2002) // Thierry Roncalli’s Home Page. URL: http://www.thierryroncalli. com/download/mixing-riskop.pdf (дата обращения: 10.02.2016)
8. Kiefer J. On Bahadur’s representation of sample quantiles // Ann. Math. Statist. 1967. Vol. 38 (5). C. 1323-1342.
9. Mignola G. Effect of a data collection threshold in the loss distribution approach / G. Mignola, R. Ugoccioni // Journal of Operational Risk. 2007. Vol. 1 (4). C. 35-47.
10. Nesˇlehova´ J. Infinite-mean models and the LDA for operational risk / J. Nesˇlehova´ , P. Embrechts, V. Chavez-Demoulin // Journal of Operational Risk. 2006. Vol. 1 (1). C. 3-25.
11. Panjer H. Operational Risk: Modeling Analytics. - Hoboken: John Wiley & Sons, 2006. 431 p.
12. The 2002 Loss Data Collection Exercise for Operational Risk: Summary of the Data Collected // Basel Committee on Banking Supervision, March 2003. URL: http://www.bis. org/bcbs/qis/ldce2002.pdf (дата обращения: 10.02.2016).
Review
For citations:
Kokh L.V., Bulatsky S.M. THE PROBLEMS OF INTERNAL AND EXTERNAL DATA USE FOR OPERATIONAL RISK ASSESSMENT IN COMMERCIAL BANKS. Issues of Risk Analysis. 2016;13(4):72-79. (In Russ.) https://doi.org/10.32686/1812-5220-2016-13-4-72-79
Views:
415
Email this article 