Programming the statistical procedures from SAS

HAS PROC REG been out of date?

Accepted Solution Solved
Reply
Frequent Contributor
Posts: 88
Accepted Solution

HAS PROC REG been out of date?

Dear All,

 

Just want to ask a question: I think PROC MIXED, PROC GLM are very powerful in the regression analysis, could I say that PROC REG has been out of date?

What is the main reason that we should still keep such procedure in the SAS package?

 

Best,

Jack.


Accepted Solutions
Solution
‎07-19-2016 09:03 AM
SAS Super FREQ
Posts: 3,630

Re: HAS PROC REG been out of date?

I would argue that you want to use the simplest procedure possible that solves your problem, not the most general. The more specific procedures have additional tests that can be more powerful.

 

I think logistic regression is the best place to see this. PROC GENMOD can fit logistic models, but PROC LOGISTIC has specific tests (such as the Hosmer-Lemeshow GIF test) and graphics (ROC curves) that are applicable in the special case of a binary response. 

 

PROC REG uses OLS regression, which is faster and gives different results than iterative MLE estimation in the generalized regression procedures.  If you have continuous variables and want to run OLS regression, PROC REG provides a direct computation.

 

To use a math analogy, you could use complex numbers to balance your checkbook, but you don't. Real numbers are simpler, easier to work with, and they get the job done.  

View solution in original post


All Replies
Respected Advisor
Posts: 4,819

Re: HAS PROC REG been out of date?

proc reg has some unique features that could be implemented in glm or mixed but I guess the real reason proc reg cannot go away is the tremendous legacy of code relying on it.

PG
Solution
‎07-19-2016 09:03 AM
SAS Super FREQ
Posts: 3,630

Re: HAS PROC REG been out of date?

I would argue that you want to use the simplest procedure possible that solves your problem, not the most general. The more specific procedures have additional tests that can be more powerful.

 

I think logistic regression is the best place to see this. PROC GENMOD can fit logistic models, but PROC LOGISTIC has specific tests (such as the Hosmer-Lemeshow GIF test) and graphics (ROC curves) that are applicable in the special case of a binary response. 

 

PROC REG uses OLS regression, which is faster and gives different results than iterative MLE estimation in the generalized regression procedures.  If you have continuous variables and want to run OLS regression, PROC REG provides a direct computation.

 

To use a math analogy, you could use complex numbers to balance your checkbook, but you don't. Real numbers are simpler, easier to work with, and they get the job done.  

Frequent Contributor
Posts: 88

Re: HAS PROC REG been out of date?

Many thanks for your quick and detailed clarification.

Respected Advisor
Posts: 2,655

Re: HAS PROC REG been out of date?

There really is no other procedure available for looking at collinearity diagnostics.  For that alone, PROC REG is worth having.

 

Steve Denham

Super User
Posts: 9,874

Re: HAS PROC REG been out of date?


And don't forget COOK-D statistic which can check which obs has the 
most contribution to model.

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 5 replies
  • 395 views
  • 9 likes
  • 5 in conversation