风险相关性下的银行非预期操作风险集成度量——基于动力学模型视角
|
摘要
依据我国商业银行1994年~2012年操作风险历史数据,建立考虑不同时点风险单元间相关性的动力学模型来描述操作风险损失产生、传导及演化的机制.该模型指出操作风险损失产生的机制有三种:自发产生的损失、不同事件类型之间的相互影响以及银行覆盖预期损失的资本金储备.通过对损失演化模型的仿真计算,获得具有低频高损特征的非预期操作风险损失情景模拟,并计算年度总体非预期损失的VaR.实证结果表明:动力学模型能够很好地描述不同风险事件类型间不同时点的相依性结构,并通过稳健性检验;在较高置信水平下,线性叠加VaR会高估风险,高估程度会随置信水平的提升而提升;内部欺诈已成为中国商业银行操作风险最主要的风险事件类型.
Based on historical operational loss data in Chinese commercial banks from year 1994 to 2012,a dynamical model containing different-time correlations is established to describe the mechanism of unexpected loss occurrence,transmission and evolution. The model takes into account the interactions among different event types,the spontaneous generation of losses and the economical capital reservation set by banks to cover expected losses. Through simulations of loss evolution model,unexpected operational risk loss scenarios( low frequency high severity) and calculation of one-year VaR of total unexpected aggregate losses can be achieved.The empirical results demonstrate that the dynamical model is good for description of different-time dependence structures among event types and can pass the robustness test. At a high confidence level,additive sum of VaR systematically overestimates total risk and the degree of overestimation will go up with rising of confidence level. Internal fraud has become a main type of operational risk in Chinese commercial banks.
Based on historical operational loss data in Chinese commercial banks from year 1994 to 2012,a dynamical model containing different-time correlations is established to describe the mechanism of unexpected loss occurrence,transmission and evolution. The model takes into account the interactions among different event types,the spontaneous generation of losses and the economical capital reservation set by banks to cover expected losses. Through simulations of loss evolution model,unexpected operational risk loss scenarios( low frequency high severity) and calculation of one-year VaR of total unexpected aggregate losses can be achieved.The empirical results demonstrate that the dynamical model is good for description of different-time dependence structures among event types and can pass the robustness test. At a high confidence level,additive sum of VaR systematically overestimates total risk and the degree of overestimation will go up with rising of confidence level. Internal fraud has become a main type of operational risk in Chinese commercial banks.
引文
[1]BaselⅡ.International Convergence of Capital Measurement and Capital Standards[R].Switzerland:Basel Committee on Banking Supervision,2004.
[2]Cruz M G.Modeling,Measuring and Hedging Operational Risk[M].New York:John Willey,2002:210-230.
[3]Chavez-Demoulin V,Embrechts P,Hofert M.An extreme value approach for modeling operational risk losses depending on covariates[J].Journal of Risk and Insurance,2016,83(3):735-776.
[4]Zhou X,Durfee A V,Fabozzi F J.On stability of operational risk estimates by LDA:From causes to approaches[J].Journal of Banking&Finance,2016,68(7):266-287.
[5]Moscadelli M.The Modelling of Operational Risk:Experience with the Analysis of the Data Collected by the Basel Committee[R].Banca d’Italia,2004.
[6]Mc Neil A,Frey R,Embrechts P.Quantitative Risk Management,Concepts,Techniques[M].New Jersey:Princeton University Press,2005.
[7]Chavez-Demoulin V,Embrechts P,Neslehova J.Quantitative models for operational risk:Extremes,dependence and aggregation[J].Journal of Banking&Finance,2006,30(10):2635-2658.
[8]Abbate D,Gourier E,Farkas W.Operational risk quantification using extreme value theory and copulas:From theory to practice[J].Social Science Electronic Publishing,2008,4(3):3-26.
[9]Brechmann E,Czado C,Paterlini S.Flexible dependence modeling of operational risk losses and its impact on total capital requirements[J].Journal of Banking&Finance,2014,40(3):271-285.
[10]Bocker K,Klüppelberg C.Modelling and measuring multivariate operational risk with Lévy copulas[J].Operational Risk,2008,3(2):3-27.
[11]Grothe O,Hofert M.Construction and sampling of Archimedean and nested Archimedean Lévy copulas[J].Journal of Multivariate Analysis,2015,138(3):182-198.
[12]Avanzi B,Tao J,Wong B,et al.Capturing non-exchangeable dependence in multivariate loss processes with nested Archimedean Lévy copulas[J].Annals of Actuarial Science,2016,10(1):87-117.
[13]Kühn R,Neu P.Functional correlation approach to operational risk in banking organizations[J].Physica A:Statistical Mechanics and Its Applications,2003,322(5):650-666.
[14]Chernobai A,Yildirim Y.The dynamics of operational loss clustering[J].Journal of Banking&Finance,2008,32(12):2655-2666.
[15]Mittnik S,Starobinskaya I.Modeling dependencies in operational risk with hybrid Bayesian networks[J].Methodology&Computing in Applied Probability,2010,12(3):379-390.
[16]Neil M,Lasse D,Andersen B,et al.Modelling operational risk in financial institutions using hybrid dynamic Bayesian networks[J].Journal of Operational Risk,2009,4(1):3-33.
[17]刘睿,詹原瑞,刘家鹏.基于贝叶斯MCMC的POT模型---低频高损的操作风险度量[J].管理科学,2007,20(3):76-83.Liu Rui,Zhan Yuanrui,Liu Jiapeng.The pot model based on Bayesian MCMC method:Measuring low frequency/high severity operational risk[J].Journal of Management Sciences,2007,20(3):76-83.(in Chinese)
[18]Lu Z.Modeling the yearly Value-at-Risk for operational risk in Chinese commercial banks[J].Mathematics and Computers in Simulation,2011,82(4):604-616.
[19]高丽君,李建平,徐伟宣,等.基于POT方法的商业银行操作风险极端值估计[J].运筹与管理,2007,16(1):112-117.Gao Lijun,Li Jianping,Xu Weixuan,et al.Estimating extreme values of operational risk for commercial banks via peak over threshold method[J].Operations Research and Management Science,2007,16(1):112-117.(in Chinese)
[20]Han J,Wang W,Wang J.POT model for operational risk:Experience with the analysis of the data collected from Chinese commercial banks[J].China Economic Review,2015,36(C):325-340.
[21]Li J,Feng J,Chen J.A piecewise-defined severity distribution-based loss distribution approach to estimate operational risk:Evidence from Chinese national commercial banks[J].International Journal of Information Technology&Decision Making,2009,8(4):727-747.
[22]张明善,唐小我,莫建明.Weibull分布下操作风险监管资本及度量精度灵敏度[J].系统工程理论与实践,2014,34(8):1932-1943.Zhang Mingshan,Tang Xiaowo,Mo Jianming.Sensitivity of operational risk’s regulatory capital and measurement accuracy in the Weibull distribution[J].Systems Engineering:Theory&Practice,2014,34(8):1932-1943.(in Chinese)
[23]Wang Z,Wang W,Chen X,et al.Using BS-PSD-LDA approach to measure operational risk of Chinese commercial banks[J].Economic Modelling,2012,29(6):2095-2103.
[24]王宗润,汪武超,陈晓红,等.基于BS抽样与分段定义损失强度操作风险度量[J].管理科学学报,2012,15(12):58-69.Wang Zongrun,Wang Wuchao,Chen Xiaohong,et al.Application of LDA based on bootstrap sampling and piecewise-defined severity distribution in operational risk measurement[J].Journal of Management Sciences in China,2012,15(12):58-69.(in Chinese)
[25]司马则茜,蔡晨,李建平.基于g-h分布度量银行操作风险[J].系统工程理论实践,2011,31(12):2321-2327.Sima Zeqian,Cai Chen,Li Jianping.Bank operational risk measurement based on g-h distribution[J].Systems Engineering:Theory&Practice,2011,31(12):2321-2327.(in Chinese)
[26]Feng J,Li J,Gao L,et al.A combination model for operational risk estimation in a Chinese banking industry case[J].Journal of Operational Risk,2012,7(2):17-39.
[27]汪冬华,徐驰.基于非参数方法的银行操作风险度量[J].管理科学学报,2015,18(3):104-113.Wang Donghua,Xu Chi.Operational risk measures for banks based on nonparametric methods[J].Journal of Management Sciences in China,2015,18(3):104-113.(in Chinese)
[28]李建平,丰吉闯,宋浩,等.风险相关性下的信用风险、市场风险和操作风险集成度量[J].中国管理科学,2010,18(1):18-25.Li Jianping,Feng Jichuang,Song Hao,et al.Integrating credit,market and operational risk based on risk correlation[J].Chinese Journal of Management Science,2010,18(1):18-25.(in Chinese)
[29]汪冬华,黄康,龚朴.我国商业银行整体风险度量及其敏感性分析---基于我国商业银行财务数据和金融市场公开数据[J].系统工程理论与实践,2013,33(2):284-295.Wang Donghua,Huang Kang,Gong Pu.Intergrated risk measurement of Chinese commercial banks and its sensitivity:Based on the financial data of Chinese commercial banks and the open data of the financial market[J].Systems Engineering:Theory&Practice,2013,33(2):284-295.(in Chinese)
[30]于文华,魏宇,康明惠.不同时变Copula-EVT-ES模型精度比较研究[J].管理科学学报,2015,18(5):32-45.Yu Wenhua,Wei Yu,Kang Minghui.Comparative study on measurement precision of different time-varying Copula-EVT-ES models[J].Journal of Management Sciences in China,2015,18(5):32-45.(in Chinese)
[31]吴吉林,陈刚,黄辰.中国A、B、H股市间尾部相依性的趋势研究---基于多机制平滑转换混合Copula模型的实证分析[J].管理科学学报,2015,18(2):50-65.Wu Jilin,Chen Gang,Huang Chen.Long-term dynamic trends in tail dependence of Chinese A,B and H stock markets:Empirical analysis based on multi-regime smoothing transition mixed copula model[J].Journal of Management Sciences in China,2015,18(2):50-65.(in Chinese)
[32]Li J,Zhu X,Xie Y,et al.The mutual-information-based variance-covariance approach:An application to operational risk aggregation in Chinese banking[J].Journal of Operational Risk,2014,9(3):3-19.
[2]Cruz M G.Modeling,Measuring and Hedging Operational Risk[M].New York:John Willey,2002:210-230.
[3]Chavez-Demoulin V,Embrechts P,Hofert M.An extreme value approach for modeling operational risk losses depending on covariates[J].Journal of Risk and Insurance,2016,83(3):735-776.
[4]Zhou X,Durfee A V,Fabozzi F J.On stability of operational risk estimates by LDA:From causes to approaches[J].Journal of Banking&Finance,2016,68(7):266-287.
[5]Moscadelli M.The Modelling of Operational Risk:Experience with the Analysis of the Data Collected by the Basel Committee[R].Banca d’Italia,2004.
[6]Mc Neil A,Frey R,Embrechts P.Quantitative Risk Management,Concepts,Techniques[M].New Jersey:Princeton University Press,2005.
[7]Chavez-Demoulin V,Embrechts P,Neslehova J.Quantitative models for operational risk:Extremes,dependence and aggregation[J].Journal of Banking&Finance,2006,30(10):2635-2658.
[8]Abbate D,Gourier E,Farkas W.Operational risk quantification using extreme value theory and copulas:From theory to practice[J].Social Science Electronic Publishing,2008,4(3):3-26.
[9]Brechmann E,Czado C,Paterlini S.Flexible dependence modeling of operational risk losses and its impact on total capital requirements[J].Journal of Banking&Finance,2014,40(3):271-285.
[10]Bocker K,Klüppelberg C.Modelling and measuring multivariate operational risk with Lévy copulas[J].Operational Risk,2008,3(2):3-27.
[11]Grothe O,Hofert M.Construction and sampling of Archimedean and nested Archimedean Lévy copulas[J].Journal of Multivariate Analysis,2015,138(3):182-198.
[12]Avanzi B,Tao J,Wong B,et al.Capturing non-exchangeable dependence in multivariate loss processes with nested Archimedean Lévy copulas[J].Annals of Actuarial Science,2016,10(1):87-117.
[13]Kühn R,Neu P.Functional correlation approach to operational risk in banking organizations[J].Physica A:Statistical Mechanics and Its Applications,2003,322(5):650-666.
[14]Chernobai A,Yildirim Y.The dynamics of operational loss clustering[J].Journal of Banking&Finance,2008,32(12):2655-2666.
[15]Mittnik S,Starobinskaya I.Modeling dependencies in operational risk with hybrid Bayesian networks[J].Methodology&Computing in Applied Probability,2010,12(3):379-390.
[16]Neil M,Lasse D,Andersen B,et al.Modelling operational risk in financial institutions using hybrid dynamic Bayesian networks[J].Journal of Operational Risk,2009,4(1):3-33.
[17]刘睿,詹原瑞,刘家鹏.基于贝叶斯MCMC的POT模型---低频高损的操作风险度量[J].管理科学,2007,20(3):76-83.Liu Rui,Zhan Yuanrui,Liu Jiapeng.The pot model based on Bayesian MCMC method:Measuring low frequency/high severity operational risk[J].Journal of Management Sciences,2007,20(3):76-83.(in Chinese)
[18]Lu Z.Modeling the yearly Value-at-Risk for operational risk in Chinese commercial banks[J].Mathematics and Computers in Simulation,2011,82(4):604-616.
[19]高丽君,李建平,徐伟宣,等.基于POT方法的商业银行操作风险极端值估计[J].运筹与管理,2007,16(1):112-117.Gao Lijun,Li Jianping,Xu Weixuan,et al.Estimating extreme values of operational risk for commercial banks via peak over threshold method[J].Operations Research and Management Science,2007,16(1):112-117.(in Chinese)
[20]Han J,Wang W,Wang J.POT model for operational risk:Experience with the analysis of the data collected from Chinese commercial banks[J].China Economic Review,2015,36(C):325-340.
[21]Li J,Feng J,Chen J.A piecewise-defined severity distribution-based loss distribution approach to estimate operational risk:Evidence from Chinese national commercial banks[J].International Journal of Information Technology&Decision Making,2009,8(4):727-747.
[22]张明善,唐小我,莫建明.Weibull分布下操作风险监管资本及度量精度灵敏度[J].系统工程理论与实践,2014,34(8):1932-1943.Zhang Mingshan,Tang Xiaowo,Mo Jianming.Sensitivity of operational risk’s regulatory capital and measurement accuracy in the Weibull distribution[J].Systems Engineering:Theory&Practice,2014,34(8):1932-1943.(in Chinese)
[23]Wang Z,Wang W,Chen X,et al.Using BS-PSD-LDA approach to measure operational risk of Chinese commercial banks[J].Economic Modelling,2012,29(6):2095-2103.
[24]王宗润,汪武超,陈晓红,等.基于BS抽样与分段定义损失强度操作风险度量[J].管理科学学报,2012,15(12):58-69.Wang Zongrun,Wang Wuchao,Chen Xiaohong,et al.Application of LDA based on bootstrap sampling and piecewise-defined severity distribution in operational risk measurement[J].Journal of Management Sciences in China,2012,15(12):58-69.(in Chinese)
[25]司马则茜,蔡晨,李建平.基于g-h分布度量银行操作风险[J].系统工程理论实践,2011,31(12):2321-2327.Sima Zeqian,Cai Chen,Li Jianping.Bank operational risk measurement based on g-h distribution[J].Systems Engineering:Theory&Practice,2011,31(12):2321-2327.(in Chinese)
[26]Feng J,Li J,Gao L,et al.A combination model for operational risk estimation in a Chinese banking industry case[J].Journal of Operational Risk,2012,7(2):17-39.
[27]汪冬华,徐驰.基于非参数方法的银行操作风险度量[J].管理科学学报,2015,18(3):104-113.Wang Donghua,Xu Chi.Operational risk measures for banks based on nonparametric methods[J].Journal of Management Sciences in China,2015,18(3):104-113.(in Chinese)
[28]李建平,丰吉闯,宋浩,等.风险相关性下的信用风险、市场风险和操作风险集成度量[J].中国管理科学,2010,18(1):18-25.Li Jianping,Feng Jichuang,Song Hao,et al.Integrating credit,market and operational risk based on risk correlation[J].Chinese Journal of Management Science,2010,18(1):18-25.(in Chinese)
[29]汪冬华,黄康,龚朴.我国商业银行整体风险度量及其敏感性分析---基于我国商业银行财务数据和金融市场公开数据[J].系统工程理论与实践,2013,33(2):284-295.Wang Donghua,Huang Kang,Gong Pu.Intergrated risk measurement of Chinese commercial banks and its sensitivity:Based on the financial data of Chinese commercial banks and the open data of the financial market[J].Systems Engineering:Theory&Practice,2013,33(2):284-295.(in Chinese)
[30]于文华,魏宇,康明惠.不同时变Copula-EVT-ES模型精度比较研究[J].管理科学学报,2015,18(5):32-45.Yu Wenhua,Wei Yu,Kang Minghui.Comparative study on measurement precision of different time-varying Copula-EVT-ES models[J].Journal of Management Sciences in China,2015,18(5):32-45.(in Chinese)
[31]吴吉林,陈刚,黄辰.中国A、B、H股市间尾部相依性的趋势研究---基于多机制平滑转换混合Copula模型的实证分析[J].管理科学学报,2015,18(2):50-65.Wu Jilin,Chen Gang,Huang Chen.Long-term dynamic trends in tail dependence of Chinese A,B and H stock markets:Empirical analysis based on multi-regime smoothing transition mixed copula model[J].Journal of Management Sciences in China,2015,18(2):50-65.(in Chinese)
[32]Li J,Zhu X,Xie Y,et al.The mutual-information-based variance-covariance approach:An application to operational risk aggregation in Chinese banking[J].Journal of Operational Risk,2014,9(3):3-19.