Programming the statistical procedures from SAS

Proc GLM or Proc Reg?

Accepted Solution Solved
Reply
Regular Contributor
Posts: 215
Accepted Solution

Proc GLM or Proc Reg?

Assume we would like to investigate if there is a suitable biomarker to detect the progressing of disease A.   We select five biomarkers and 30 participants.  We record the different periods of those biomarker concentrations, which are 0 month, 6 months, 12 months and 24 months.  And the golden standard diagnosis is X-ray ( Yes/No). 

To find the best biomarker, what statistical model I should use?  Multilinear regression?  How to write the SAS code?  Please help.  Thanks.


Accepted Solutions
Solution
‎01-19-2017 11:15 AM
Grand Advisor
Posts: 16,893

Re: Proc GLM or Proc Reg?

The GLM repeated measures example in the documentation is close to what you need. 

 

You can't use PROC REG because you have a repeated measures design, which requires PROC GLM or PROC MIXED. 

 

Full methodological support is a bit outside the scope of this forum IMO. 

If you're not sure what methods to be using consider taking the FREE SAS statistical course. 

See bottom right hand corner of the page. 

http://support.sas.com/training/tutorial/

 

FYI 30 participants is not enough to determine conclusive results. 

View solution in original post


All Replies
Solution
‎01-19-2017 11:15 AM
Grand Advisor
Posts: 16,893

Re: Proc GLM or Proc Reg?

The GLM repeated measures example in the documentation is close to what you need. 

 

You can't use PROC REG because you have a repeated measures design, which requires PROC GLM or PROC MIXED. 

 

Full methodological support is a bit outside the scope of this forum IMO. 

If you're not sure what methods to be using consider taking the FREE SAS statistical course. 

See bottom right hand corner of the page. 

http://support.sas.com/training/tutorial/

 

FYI 30 participants is not enough to determine conclusive results. 

☑ This topic is SOLVED.

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

Discussion stats
  • 1 reply
  • 86 views
  • 0 likes
  • 2 in conversation