An Idea Exchange for SAS software and services

by Respected Advisor
on ‎12-12-2012 11:05 AM
Hi! Participating in a conference on invasive algae. What can I do for you?
by Super Contributor
on ‎12-12-2012 11:12 AM
Thank you! I need to Google "algae".

I wander if I give you my office number, would you please leave me a message ? both in French and English. Say simple things like:
Hi Linlin, this is Pierre Gagnon, how are you doing? what are you going to do today?
by Respected Advisor
on ‎12-12-2012 11:30 AM
OK Im using my phone for the conference right now. I'll do it in the afternoon. Why? You just want to hear my voice?
by Super Contributor
on ‎12-12-2012 11:40 AM
Thank you!!! Yes. I will email you my number at the end of the day.
by PROC Star
on ‎01-26-2014 11:02 AM

My vote isn't just for more and consistent functionality, but for both expanded applicability (not just in SAS/Stat) and output consistent with by statements (just not requiring pre-sorting).

by Frequent Contributor
on ‎01-26-2014 11:10 AM

I didn't even know there WAS a CLASS statement outside of SAS/Stat. But definitely consistent output is good.

by PROC Star
on ‎01-26-2014 02:18 PM

Peter,  surely you've heard of proc means/summary and, no, they're not part of SAS/Stat!

by Frequent Contributor
on ‎01-26-2014 02:25 PM

(slap self on head) Oh, yeah. I forgot those were in BASE!

by Valued Guide
on ‎01-29-2014 09:22 AM

Going back to the introduction of PROC GLM in the 1970s, SAS has emphasized the advantages of the so-called "GLM" parameterization for factorials (what you get with a CLASS statement in MIXED, for instance). It has to do with Type I through Type IV testable hypotheses and unbalanced data. I doubt if SAS will add other parameterizations to the CLASS statement in MIXED, GLIMMIX, etc.

by Respected Advisor
on ‎01-08-2016 01:00 PM

I have to agree with Larry on this one--GLM and MIXED would tend to puke all over themselves with a REF type definition.

Idea Statuses
Top Liked Authors