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

徐威廉 · 2024年11月02日

C选项

NO.PZ2024042601000018

问题如下:

The head of quantitative analytics (HQA) of a bank is evaluating a proposal to change the bank’s default prediction process from predominantly relying on ratings from rating agencies to applying internal rating models that rely on statistical-based and structural approaches. As part of the proposal, the HQA needs to justify to the senior management the advantages of using these structural models for default prediction. Which of the following observations would the HQA be correct to make?

选项:

A.

When applying logistic regression models, the coefficients are estimated by the maximum likelihood estimation (MLE) method.

B.

The Merton model can be effectively applied to both private and publicly listed companies.

C.

The coefficients of linear discriminant analysis are estimated by the method of least squares.

D.

The KMV model includes equity and optionless long-term debt only, which can simplify the analysis of issuers with complex capital structures.

解释:

A is correct. In logistic regression, coefficients are estimated by using the ’maximum likelihood estimation’ (MLE) method.

B is incorrect. The Merton model has several limitations. It is applicable to liquid, publicly traded names only, and there is continuous need for calibration, which is costly for smaller banks. The Merton model relies on the continually changing movements in market prices, volatility, and interest rates.

C is incorrect. The coefficients of LDA are estimated by (1) maximizing the homogeneity around the two centroids and (2) minimizing the overlapping zone in which the two groups of borrowers are mixed and share similar Z-scores.

D is incorrect. The KMV model is a dynamic model that can be applied to issuers with complex capital structures. Equity, short-term debt, long-term debt, convertible debt are all included in the analysis of default prediction.

C选项错哪了?能不能把A选项和C选项解释一下

1 个答案
已采纳答案

pzqa27 · 2024年11月04日

嗨,从没放弃的小努力你好:


A的 logistic regression models 在一级学过,它的斜率就是用极大似然估计法来做的。

C说判别模型的斜率是用least squares的方法来估计的,判别模型典型代表是Z score。它主要通过计算出一个score来判断下公司是否处于违约的门槛内,因此它的斜率是由2个要求,就是解析写的那2个(1) maximizing the homogeneity around the two centroids and (2) minimizing the overlapping zone in which the two groups of borrowers are mixed and share similar Z-scores.

----------------------------------------------
虽然现在很辛苦,但努力过的感觉真的很好,加油!

  • 1

    回答
  • 0

    关注
  • 28

    浏览
相关问题