重尾性操作风险的风险价值置信区间的灵敏度

Confidence intervals' sensitivity of heavy-tailed operational VaR

根据操作风险价值不确定度的合成机理,通过对重尾性操作风险价值解析解的分析,导出高置信度下重尾性操作风险价值的标准不确定度及置信区间.进而以弹性分析方法对该置信区间的灵敏度进行理论探讨和实例分析后发现:随尾指数的增大,操作风险价值及其置信区间长度同时增大,且在高置信度下,尾指数是影响操作风险价值及其置信区间长度灵敏度的关键参数.据此,可将尾指数作为操作风险的监控参数,操作风险监管资本的提取方式可改进为:以操作风险价值的某一置信区间为监管资本的控制范围.这在理论上进一步完善了损失分布法在操作风险度量中的应用,并使操作风险的管理更加合理.

Based on the synthesis mechanism of the operational VaR＇s uncertainty, this paper makes an analysis of the solutions of heavy-tailed operational VaR. And the standard uncertainty and the confidence intervals of operational VaR are obtained at high confidence levels. By the theory research and the numerical example analysis of confidence intervals＇ sensitivity on the basis of elasticity analysis method, a rule is discovered that the operational VaR and the confidence intervals＇ length increase with the tail-index, and the tail-index is a key parameter that influences the operational VaR and the confidence intervals＇ length at high confidence levels. Accordingly, the tail-index can be regarded as a parameter of monitoring the operational risk, and the extraction method of regulatory capital can be mended that the regulatory capital is controlled in a certain confidence interval of operational VaR. This research improves the application of the loss distribution approach to the operational risk measurement, and makes the management of operational risk more reasonable.