FRM®知识点:Model verification based on failure rates
文章来源:网络
发布时间:2022-01-26 14:29
阅读:626次

When T is large,we can use the central limit theorem and approximate the binomial distribution by the normal distribution.
Example:
In its 1998 annual report,J.P.Morgan explained that
in 1998,daily revenue fell short of the downside(95%VaR)band…on 20 days,or more than 5%of the time.Nine of these 20 occurrences fell within the Augest to October period.
We can test wheher this was bad luck or a faulty model,assuming 252 days in the year.
We reject the hypothesis that the VaR model is unbiased.It is unlikely(at 95%test CL)that this was bad luck
Answer:【点击免费下载>>>更多FRM学习相关资料】
超过95%VaR的天数为20天,20/250>95%,在8月至9月中就发生了9天。
原假设:p=0.05 x=20
拒绝原假设说明模型有问题。
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

Lucy

FRM持证人,CFA持证人,CICPA会员(注册会计师),ACCA会员,国际注册内部审计师(CIA),17年财务从业者,从事金融证书培训5年有余,授课紧跟当下热点话题,能迅速与00后学员找到共同话题引起共鸣,直播幽默有趣,被学员亲切的称为“女明星”。上课逻辑清晰,有条理,能帮助学员快乐学习的同时又能迅速掌握知识点。
