Operational Risk Measurement: A Loss Distribution Approach with Segmented Dependence

IF 0.4 4区经济学 Q4 BUSINESS, FINANCE

Journal of Operational Risk Pub Date : 2019-02-11 DOI:10.21314/JOP.2019.220

Xiaoqian Zhu, Yinghui Wang, Jianping Li

{"title":"Operational Risk Measurement: A Loss Distribution Approach with Segmented Dependence","authors":"Xiaoqian Zhu, Yinghui Wang, Jianping Li","doi":"10.21314/JOP.2019.220","DOIUrl":null,"url":null,"abstract":"In the loss distribution approach (LDA), the most widely used approach of operational risk measurement, the modeling dependencies across different risk cells have been extensively studied. However, it has not been recognized that the dependencies between high-frequency, low-impact (HFLI) and low-frequency, high-impact (LFHI) operational risk losses are naturally different. This paper proposes an approach, called the loss distribution approach with segmented dependence (LDA-SD), which can model the different dependencies of HFLI and LFHI losses in the framework of LDA. LDA-SD divides the losses into two parts for HFLI and LFHI, fits their frequency and severity distributions separately and models the segmented dependencies with a copula. In our empirical study, the proposed LDA-SD is applied to measure the operational risk of the overall Chinese banking sector based on the Chinese Operational Loss Database data set, the largest operational risk data set in China. The empirical results reveal that the dependencies are indeed different between HFLI and LFHI losses. The operational risk capital calculated by the LDA-SD is significantly smaller than that calculated by the LDA and considering the holistic dependence, but larger than that simply considering tail dependence.","PeriodicalId":54030,"journal":{"name":"Journal of Operational Risk","volume":"11 1","pages":""},"PeriodicalIF":0.4000,"publicationDate":"2019-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"15","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Operational Risk","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.21314/JOP.2019.220","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BUSINESS, FINANCE","Score":null,"Total":0}

引用次数: 15

Abstract

In the loss distribution approach (LDA), the most widely used approach of operational risk measurement, the modeling dependencies across different risk cells have been extensively studied. However, it has not been recognized that the dependencies between high-frequency, low-impact (HFLI) and low-frequency, high-impact (LFHI) operational risk losses are naturally different. This paper proposes an approach, called the loss distribution approach with segmented dependence (LDA-SD), which can model the different dependencies of HFLI and LFHI losses in the framework of LDA. LDA-SD divides the losses into two parts for HFLI and LFHI, fits their frequency and severity distributions separately and models the segmented dependencies with a copula. In our empirical study, the proposed LDA-SD is applied to measure the operational risk of the overall Chinese banking sector based on the Chinese Operational Loss Database data set, the largest operational risk data set in China. The empirical results reveal that the dependencies are indeed different between HFLI and LFHI losses. The operational risk capital calculated by the LDA-SD is significantly smaller than that calculated by the LDA and considering the holistic dependence, but larger than that simply considering tail dependence.

查看原文本刊更多论文

操作风险度量:一种具有分段依赖的损失分配方法

在应用最广泛的操作风险度量方法损失分布法(LDA)中，对不同风险单元之间的建模依赖关系进行了广泛的研究。然而，人们还没有认识到，高频、低影响(HFLI)和低频、高影响(LFHI)操作风险损失之间的依赖关系自然是不同的。本文提出了一种分段依赖损失分布方法(LDA- sd)，该方法可以在LDA框架下对HFLI和LFHI损失的不同依赖关系进行建模。LDA-SD将HFLI和LFHI的损失分为两部分，分别拟合它们的频率和严重程度分布，并用copula对分割后的依赖关系进行建模。在我们的实证研究中，我们基于中国最大的操作风险数据集——中国操作损失数据库数据集，将提出的LDA-SD用于衡量中国银行业的整体操作风险。实证结果表明，HFLI和LFHI损失之间的依赖关系确实不同。LDA- sd计算的操作风险资本明显小于考虑整体依赖的LDA计算的操作风险资本，但大于单纯考虑尾部依赖的操作风险资本。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Operational Risk BUSINESS, FINANCE-

CiteScore

1.00

自引率

40.00%

发文量

期刊介绍： In December 2017, the Basel Committee published the final version of its standardized measurement approach (SMA) methodology, which will replace the approaches set out in Basel II (ie, the simpler standardized approaches and advanced measurement approach (AMA) that allowed use of internal models) from January 1, 2022. Independently of the Basel III rules, in order to manage and mitigate risks, they still need to be measurable by anyone. The operational risk industry needs to keep that in mind. While the purpose of the now defunct AMA was to find out the level of regulatory capital to protect a firm against operational risks, we still can – and should – use models to estimate operational risk economic capital. Without these, the task of managing and mitigating capital would be incredibly difficult. These internal models are now unshackled from regulatory requirements and can be optimized for managing the daily risks to which financial institutions are exposed. In addition, operational risk models can and should be used for stress tests and Comprehensive Capital Analysis and Review (CCAR). The Journal of Operational Risk also welcomes papers on nonfinancial risks as well as topics including, but not limited to, the following. The modeling and management of operational risk. Recent advances in techniques used to model operational risk, eg, copulas, correlation, aggregate loss distributions, Bayesian methods and extreme value theory. The pricing and hedging of operational risk and/or any risk transfer techniques. Data modeling external loss data, business control factors and scenario analysis. Models used to aggregate different types of data. Causal models that link key risk indicators and macroeconomic factors to operational losses. Regulatory issues, such as Basel II or any other local regulatory issue. Enterprise risk management. Cyber risk. Big data.