开发者:上海品职教育科技有限公司 隐私政策详情

应用版本:4.2.11(IOS)｜3.2.5(安卓)APP下载

返回

vivian_zm · 2020年02月24日

问一道题：NO.PZ2019100902000019

  1. FRM II
  2. Credit Risk Measurement And Management

问题如下：

which of the following statements is incorrect?

选项：

A.
In the cumulative accuracy profile or CAP(also called GINI curve), the accuracy ratio (AR) is defined as AR/AR+Ap, the closer this ratio is to 1, the more accurate is the model.
B.

K-nearest neighbor approach uses Non-parametric classification

C.

in the Support vector machines, the main idea is to separate the various classes of data (observations) using the best separating hyperplane (BSH), for which the distance to two parallel hyperplanes on each side is the largest.

D.
In the Fisher Linear discriminant analysis, the principal aim is to segregate and classify a heterogeneous population in heterogeneous subsets.

解释：

答案：D is correct.

解析：本题让选项错误的一个描述，A/B/C三个描述均正确。D选项错误，D选项的正确描述为：

the principal aim is to segregate and classify a heterogeneous population in homogeneous subsets.

请问能否解释一下A选项啥意思呀？好像学到这里没有涉及?谢谢！

添加评论
1 个答案
已采纳答案

orange品职答疑助手 · 2020年02月25日

同学你好，这个知识点在credit rating这一章里。perfect model是指风险最高的前5%公司，他们的真实违约概率都是100%。所以perfect model预测得飞常准。如果AR/AR+AP越大，说明actual rating model就越接近perfect model。

具体的你也可以听一听基础班这个知识点的课

  • 1

    回答

  • 0

    关注

  • 409

    浏览
相关问题

NO.PZ2019100902000019 K-nearest neighbor approauses Non-parametric classification in the Support vector machines, the main ia is to separate the various classes of ta (observations) using the best separating hyperplane (BSH), for whithe stanto two parallel hyperplanes on easi is the largest. In the Fisher Linescriminant analysis, the principaim is to segregate anclassify a heterogeneous population in heterogeneous subsets. 答案is correct. 解析本题让错误的一个描述，A/B/C三个描述均正确。误，正确描述为 the principaim is to segregate anclassify a heterogeneous population in homogeneous subsets. 不太理解c的意思

2021-03-27 10:48 1 · 回答

NO.PZ2019100902000019 1.AR值(AccuraRatio)在原版书多少页有，没找着。 2.公式里的AP是什么意思，AR又是什么意思。 3.然后还有就是。。感觉原版书和课件几乎完全对不上，像这个线性判别+K临近，还有很多定性定量跟原版书的定性定量说的也不一样，Z-score我也没找到，这个是原版书的问题吗，还是什么问题 辛苦老师解答~

2021-03-03 23:50 1 · 回答

K-nearest neighbor approauses Non-parametric classification in the Support vector machines, the main ia is to separate the various classes of ta (observations) using the best separating hyperplane (BSH), for whithe stanto two parallel hyperplanes on easi is the largest. In the Fisher Linescriminant analysis, the principaim is to segregate anclassify a heterogeneous population in heterogeneous subsets. 答案is correct. 解析本题让错误的一个描述，A/B/C三个描述均正确。误，正确描述为 the principaim is to segregate anclassify a heterogeneous population in homogeneous subsets. 老师，什么意思？以及答案的翻译。

2020-05-23 06:12 1 · 回答

K-nearest neighbor approauses Non-parametric classification in the Support vector machines, the main ia is to separate the various classes of ta (observations) using the best separating hyperplane (BSH), for whithe stanto two parallel hyperplanes on easi is the largest. In the Fisher Linescriminant analysis, the principaim is to segregate anclassify a heterogeneous population in heterogeneous subsets. 答案is correct. 解析本题让错误的一个描述，A/B/C三个描述均正确。误，正确描述为 the principaim is to segregate anclassify a heterogeneous population in homogeneous subsets. A也是错的吧，原版书322页说Accuraratio是Ar/Ap

2020-03-05 16:02 1 · 回答