Programming the statistical procedures from SAS

Requesting guidance on usage of SAS STAT procedures?

Reply
PROC Star
Posts: 1,805

Requesting guidance on usage of SAS STAT procedures?

Good afternoon Folks, I'm about to begin to my so called formal journey on using SAS stat from my past experience of data processing using the language. Therefore, I seek help in some tips to get me started that can point me where to look for what.

For example, for descriptive more than one proc such as means and summary, uni variate  may accomplish the same task. What I really need is which one to use? why? what makes it more robust or precise than the other or perhaps convenience?

 

And i guess the same perhaps  glm, reg and others but i am not diving into multivariate just yet and hence in italics

 

My statistics journey begins starting from very basic descriptive s this quarter up to time series and forecasting scheduled to be covered over the next 3 quarters. I would sincerely appreciate your  time for the valuable guidance you could offer?

 

Have a nice long weekend. Cheers!

Respected Advisor
Posts: 3,018

Re: Requesting guidance on usage of SAS STAT procedures?

Posted in reply to novinosrin

A small list of the differences:

 

Univariate gives you plots related to the distribution of the variables. MEANS/SUMMARY do not.

 

MEANS/SUMMARY give you the ability to slice and dice the analysis by any number of class variables, and come up with statistics in all of the slices. The CLASS statement in UNIVARIATE limits you to slicing the data by 2 class variables.

 

For many purposes, any of them will do the job.

 

Regarding REG, GLM, GLIMMIX and others, SAS has written a comparison of them, it's in the docs somewhere, if you can't find it let me know.

--
Paige Miller
Ask a Question
Discussion stats
  • 1 reply
  • 109 views
  • 2 likes
  • 2 in conversation