SAS Exam A00-240 Topic 2 Question 97 Discussion

Actual exam question for SAS's A00-240 exam

Question #: 97
Topic #: 2

An analyst fits a logistic regression model to predict whether or not a client will default on a loan. One of the predictors in the model is agent, and each agent serves 15-20 clients each. The model fails to converge. The analyst prints the summarized data, showing the number of defaulted loans per agent. See the partial output below:

What is the most likely reason that the model fails to converge?

AThere is quasi-complete separation in the data.

BThere is collinearity among the predictors.

CThere are missing values in the data.

DThere are too many observations in the data.

Show Suggested Answer

Suggested Answer: C

by Sabrina at Sep 19, 2024, 07:28 AM

Limited Time Offer

25%

Off

Get Premium A00-240 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Adria

25 days ago

Ah, the old 'model fails to converge' problem. Sounds like a job for the Convergence Fairy! Just sprinkle some magical convergence dust on the data, and voila - problem solved!

upvoted 0 times

...

Delmy

28 days ago

Missing values? In a certification exam question? Now that's just cruel. No, the answer has to be option A - this is a classic case of quasi-complete separation, and the model just can't handle it.

upvoted 0 times

Cassi

19 days ago

I agree, it's definitely quasi-complete separation. The model can't handle it.

upvoted 0 times

...

Selma

1 months ago

Too many observations? Hah, as if that's ever a problem in the real world! Nah, this one's got to be about the quasi-complete separation. Option A is the clear winner.

upvoted 0 times

Patria

18 days ago

I agree, quasi-complete separation is a common issue in logistic regression models.

upvoted 0 times

...

Naomi

2 months ago

Hmm, I'm not sure about this one. It could be an issue with collinearity, but the problem seems more likely to be with the separation in the data. I'd go with option A.

upvoted 0 times

Heidy

28 days ago

I disagree, I believe it's due to quasi-complete separation in the data.

upvoted 0 times

...

Aleshia

1 months ago

I think it's because of collinearity among the predictors.

upvoted 0 times

...

Gaston

2 months ago

Aha! The data shows clear signs of quasi-complete separation - some agents have no defaulted loans, while others have a high proportion. This is likely causing the model to struggle to converge. Option A is the way to go here.

upvoted 0 times

...