Proc GLIMMIX estimation errors

[rsides17]

I'm running a proc glimmix model where the model estimates are constistently off (in the same direction) from the raw rates. Are there good suggestions as to how to fix this or what could be happening?

Thanks!

[ballardw]

Are you treating missing values as 0 in either set of calculations?

You might provide a little example data that will duplicate the behavior, the glimmix code and how you calculate the raw rates. Otherwise lots of guessing may ensue.

Instructions here: https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-dat... will show how to turn an existing SAS data set into data step code that can be pasted into a forum code box using the {i} icon or attached as text to show exactly what you have and that we can test code against.

[sld]

If you are using a distribution other than the normal, you likely are observing the differences between the "marginal mean" and the "binomial p" as illustrated in Figure 2 and discussed in

https://dl.sciencesocieties.org/publications/aj/abstracts/107/2/811

The concept applies for distributions other than binomial.