R square in proc mixed?

[Miracle]

Hi all.

How can I obtain R square in proc mixed or maybe something similar such that I can use in order to know how much variability is explained by the mixed model?

Have a good day. Many thanks in advance.

[Dale]

I responded to this thread on the SAS-L listserv. Rather than repeating that here, I'll just post a link to my response on SAS-L.

http://listserv.uga.edu/cgi-bin/wa?A2=ind0911c&L=sas-l&F=&S=&P=37753

Dale

[Cwinds]

I'm trying to access the response to this question as I have the same one, however, the link does not work.  Could someone answer it here or help direct to the answer?  Thanks.

[lvm]

There have been several discussions of this on this site. There is no unambiguous or unique R2 where then are multiple random effects. But there are some possible metrics that have been used.

[SteveDenham]

Here is a more recent link to Dale McLerran's post in the SAS-L archives:

https://listserv.uga.edu/cgi-bin/wa?A2=ind0911C&L=SAS-L&P=R23346&1=SAS-L&9=A&I=-3&J=on&d=No+Match%3B...

Steve Denham

[TravisTCU]

Thanks for posting this link, Steve.  It seems very helpful.  I'm a novice SAS user; could provide some instruction on how to constrain the random effect variance for the intercept only model to be the same as the variance which you observe when

fitting the full model, as instructed in the link?

[SteveDenham · Accepted solution]

Run the full model and get the variance component estimates.  Then run the model again (with no fixed effects, which is the intercept only model), but specify the variance component(s) in a PARMS statement, and use the hold= option to fix them at the same values.

Steve Denham