Due to the inherent characteristics of operational loss and the data collection problem, the internal data is always insufficient and hardly to get correct and robust estimation. External data is the most recommended supplement data to internal data, but it has inherent biases, and there should be inevitable result deviation if directly mixed with internal data. How to combine external data with internal data is a hard problem and needs discussion. Since the operational data of Chinese commercial banks is scarce, the selection of candidate factors for factor method subjectively, the Bayesian method with strong subjective and the quantile methods assumptions strictly, these application all are limited to Chinese commercial banks. The factors are just splitred as the common component and the idiosyncratic component, and the loss affected by the common factors are equal. Using Macarthur's homogeneity measure, the homogeneity of the internal data and external data is estimated, which enables us to get the idiosyncratic factor of internal and external data. Then the external data is modified using scaling model and combined with the internal data. Since common model can't fit the operational risk well, while the extreme theory model can just modify the tail distribution well, the two-phase mixture model is used to fit the whole operational risk severity distribution. The threshold is set as a parameter is used to be estimated. Considering the continuity constraint at the threshold, it is found that the distribution with log-normal as the body and the generalized Pareto distribution as the tail fits well to the mixture data. With the frequency simulated, the annual loss distribution is gotten. The external data comes from the 10 years collection of our team and the internal data from the bank. The result shows that the external data should be modified before combined to internal data. The threshold selection is more stable and the result more reliable of the mixed data and mixture models. The external data modified method we used has no assumption to distribution similarity, and it gives a reference to the mixture data literature.
GAO Li-jun, GAO Xiang
. Analysis Based on Mixing Data Operational Risk with Mixture Parametric Models[J]. Chinese Journal of Management Science, 2017
, 25(5)
: 11
-16
.
DOI: 10.16381/j.cnki.issn1003-207x.2017.05.002
[1] Ganegoda A, Evans J. A scaling model for severity of operational losses using generalized additive models for location scale and shape (GAMLSS)[J], Annals of Actuarial Science, 2013,7(1):61-100
[2] Ergashev B A. A theoretical framework for incorporating scenarios into operational risk modeling[J]. Journal of Financial Services Research, 2012,41(3)1:45-161.
[3] Dutta K K, Babbel D F. Scenario analysis in the measurement of operational risk capital:achange of measure approach[J]. Journal of Risk and Insurance, 2014,81(2):303-334.
[4] Bolancé C, Guillén M, Gustafsson J, et al.Adding prior knowledge to quantitative operational risk models[J]. Journal of Operational Risk, 2013,8(1):17-32.
[5] Basel Committee on Banking Supervision, Operational risk supervisory guidelines for the advanced measurement approaches[R]. Basel,Switzerland:Bank for International Settlements, 2011.
[6] 高丽君. 商业银行操作风险外部数据的内生偏差研究[J]. 管理评论, 2011,23(07), 138-142,148.
[7] Shih J, Samad-Khan A, Medapa P. Is the size of an operational loss related to firm size?[J].Operational Risk, 2000,2(1):21-22.
[8] Wei Ran. Quantification of operational losses using firm-specific information and external database[J]. Journal of Operational Risk, 2007,1(4):3-34.
[9] Na H S, Van Den Berg J, Miranda L, et al. An econometric model to scale operational losses[J]. The Journal of Operational Risk, 2006,1(2):11-31.
[10] Lambrigger D D, Shevchenko P V, Wuthrich M V.The quantication of operational risk using internal data, relevant external data and expert opinion[J]. Journal of Operational Risk, 2007,2(3):3-27.
[11] Ergashev B, Mittnik S, Sekeris E.A Bayesian approach to extreme value estimation in operational risk modeling[J]. The Journal of Operational Risk, 2013,8(4):55-81.
[12] Cope E, Labbi A. Operational loss scaling by exposure indicators:Evidence from the ORX database[J]. Journal of Operational Risk, 2008,3(4):55-81.
[13] Shevchenko P V. Modelling operational risk using bayesian inference[M]. Berlin-Heidellerg:Springer, 2011.
[14] 徐润南. 操作风险模型的数据与建模[J]. 上海投资, 2006,(3):29-35.
[15] 吴博. 操作风险高级计量法四类数据元素的整合和应用[J]. 新金融, 2012,(7):23-27.
[16] 卢安文,任玉珑,唐浩阳.基于贝叶斯推断的操作风险度量模型研究[J].系统工程学报,2009,24(3):276-292,349.
[17] 陆静, 张佳. 基于极值理论和多元Copula函数的商业银行操作风险计量研究[J]. 中国管理科学, 2013,21(3):11-19.
[18] 周艳菊, 彭俊, 王宗润. 基于Bayesian-Copula方法的商业银行操作风险度量[J]. 中国管理科学, 2011,19(4):17-25.
[19] 司马则茜, 蔡晨, 李建平. 度量银行操作风险的POT幂率模型及其应用[J]. 2009,17(1):36-41.
[20] Dahen H, Dionne G. Scaling models for the severity and frequency of externaloperational loss data[J]. Journal of Banking and Finance, 2010,34(7):1484-1496.
[21] Jost L. Partitioning diversity into independent alpha and beta components[J]. Ecology, 2007,88(10):2427-2439.
[22] Scarrott C J, MacDonald A E. A review of extreme value threshold estimation and uncertainty quantification[J]. REVSTAT Statistical Journal, 2012,10(1):33-60.
[23] Lee D, Li W K, Wong T S T. Modeling insurance claims via a mixture exponential model combined with peaks-over-threshold approach[J]. Insurance:Mathematics and Economic, 2012,51(3):538-550.
[24] Carreau J, Bengio Y. A hybrid Pareto model for asymmetric fat-tailed data:The univariate case[J]. Extremes, 2008,12(1):53-76.
