FRM®二级知识点:Unconditional and Conditional Coverage
文章来源:网络
发布时间:2022-01-26 14:24
阅读:585次

Unconditional coverage:
we are not concerned about
independence of exception observations
or the timing of when the exceptions occur.
we simply are concerned with
the number of total exceptions.
Kupiec(1995)determined a measure to accept or reject models using the tail points of a log-likelihood ratio(LR).
We would reject the hypothesis that the model is correct if the LRuc>3.84
In the JPM example,LRuc=3.91,therefore,we reject unconditional coverage.
LRuc>3.84,则model不正确。
Conditional coverage:
by including a measure of the independence of exceptions,we can measure conditional coverage of the model.
The test for conditional coverage should be performed when exceptions are clustered together.
当exception扎堆时,用条件覆盖去检验。
如果短时间内发生的exception过多,则市场风险增加,因为VaR短时间内反应慢,还没有进行预警。
推荐阅读:【FRM®二级知识点:VaR mapping】
2022年FRM®最新学习资料包
请大家认真填写以下信息,获取2025年FRM®学习资料包,会以网盘链接的形式给到大家,点击免费领取后请尽快保存。
*姓名不能为空
*手机号错误
*验证码错误

Alex

FRM持证人,CICPA会员,立信(西安)会计师事务所高级合伙人,Bilibili知名财经up主,七年审计工作经验,执教超五年。上课专业性高,学术性强,逻辑严谨,框架清晰,在给学生教学的过程中又能输出一些有趣的行业知识,提升应试能力的同时也能给象牙塔的学生带来一些职场行业知识。执教以来成功让数十名学员获取理想成绩。赋能他们的职场和人生。
