Dear Help Community,
I'm trying to understand how to interpret my estimates in the table below. I'm using GENMOD with a NEGBIN model and log link function. Are my estimates transformed, and are they log10 or natural log transformed? How can I get my response variable back to my original unit? FYI, the response is colony forming units of bacteria.
proc genmod data=liver;
class Sex Diet;
model CFU=Sex|Diet/ Link=log Dist=NEGBIN type1 type3;
LSmeans Sex|Diet/Lines;
run;
Diet Least Squares Means | ||||
Diet | Estimate | Standard Error | z Value | Pr > |z| |
A | 6.8394 | 0.01524 | 448.84 | <.0001 |
B | 9.7593 | 0.002296 | 4249.98 | <.0001 |
C | 8.5384 | 0.02500 | 341.52 | <.0001 |
D | 8.4607 | 0.003765 | 2247.04 | <.0001 |
E | 14.9283 | 0.000149 | 100409 | <.0001 |
The estimates are on the natural log scale. To get values on the original scale, use the ILINK option in the LSMEANS statement.
Steve Denham
Steve,
Perfecto! Thanks for the natural log confimation and ILINK trick.
Available on demand!
Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.