topic Backtransformation of the means in SAS Programming

Backtransformation of the means

steph_t — Wed, 26 Apr 2017 18:31:56 GMT

I am currently using SAS studio and am wondering whether it is possible to back-transform means and standard errors (and obtain LS means table) after a sqrt transformation.

I know you can use the proc glimmix to do this with log transformed data but is there any way to do this for data transformed differently?

I have used the following code for log backtransformation

Proc glimmix data=file1;
class ANIMAL TREAT;
model NH= TREAT/ link=LOGN;
run;

proc glimmix data=file100;
CLASS ANIMAL TREAT;
MODEL NH= TREAT;
lsmeans treat / cl ilink;
ods select lsmeans;
run;

Thanks!

Re: Backtransformation of the means

art297 — Wed, 26 Apr 2017 19:46:53 GMT

I'm not well (i.e., not-at-all) versed in glmmix, but can't you just use the power(.5) setting with the link option?

Art, CEO, AnalystFinder.com

Re: Backtransformation of the means

steph_t — Thu, 27 Apr 2017 11:13:40 GMT

No the ilink function is only for the natural log transformation

Re: Backtransformation of the means

art297 — Thu, 27 Apr 2017 11:44:40 GMT

I found the following from an older post that suggests otherwise:

"On SAS-L, I replied, and since folks don't always read both, I tried to recover it from the archive:

This looks like an opportunity to use PROC GLIMMIX, and use of the LINK option. For depth LINK=LOG, and for mass LINK=POWER(0.5). Then in the LSMEANS statement, use the ILINK option, and the final values will include the estimates and their standard errors on both the transformed and original scale. The documentation reports that the standard errors on the inverse lined scale are computed by the delta method."

see: https://communities.sas.com/t5/SAS-Statistical-Procedures/How-to-back-transform-LSMEAN-standard-errors-from-PROC-MIXED/td-p/34301

Art, CEO, AnalystFinder.com