Programming the statistical procedures from SAS

PROC LOGISTIC, SELECTION=BACKWARD Variable Selection

Reply
Occasional Contributor
Posts: 12

PROC LOGISTIC, SELECTION=BACKWARD Variable Selection

I've been looking at the sample "SAS Statistical Business Analysis" exam questions at http://support.sas.com/certify/creds/samples/

 

Question #6 is as follows:

 

Question 6
When selecting variables or effects using SELECTION=BACKWARD in the LOGISTIC procedure, the business analyst's model selection terminated at Step 3.

What happened between Step 1 and Step 2?
  1. DF increased.
  2. AIC increased.
  3. Pr > Chisq increased.
  4. - 2 Log L increased.
correct_answer = "D"

 

My question is: Shouldn't the answer be "C"? My understanding is that variable selection is based on a variable's p-value. The backward selection process, by default, eliminates variables one-by-one which don't meet the 0.10 criterion to "stay" in the model.

 

Could someone please clarify if my understanding is wrong? If so, please explain why "D" is the correct answer.

 

Thanks.

Contributor
Posts: 66

Re: PROC LOGISTIC, SELECTION=BACKWARD Variable Selection

The p values of all variables in the model will change when one is removed. One or more of the remaining might become "significant".

 

The question specifically asks about the step before variable selection. As variables are removed, - 2 Log L will increase. So the suggested answer is correct.

 

Norman.

Norman.
SAS 9.4 (TS1M4) X64_7PRO WIN 6.1.7601 Service Pack 1 Workstation

Super User
Posts: 9,775

Re: PROC LOGISTIC, SELECTION=BACKWARD Variable Selection

[ Edited ]

Oh, Contra . 

Pr > Chisq( H0: beta = 0 ) is smaller is better (which means model is more significant. i.e. model can explain more variance of Y )

- 2 Log L is bigger is better(which is likelihood of X. more is better model).

Contributor
Posts: 66

Re: PROC LOGISTIC, SELECTION=BACKWARD Variable Selection

Bigger is not better.

 

The more variables that are included, the better the model fit (and the lower the deviance). However, the idea is to construct a model with only as many variables as are required.

 

Removing variables will decrease the model fit, but this, in itself, is not a bad thing.

 

Norman.

Norman.
SAS 9.4 (TS1M4) X64_7PRO WIN 6.1.7601 Service Pack 1 Workstation

Ask a Question
Discussion stats
  • 3 replies
  • 372 views
  • 0 likes
  • 3 in conversation