ABE using GLM

Phoenix Certara Forums Phoenix WNL basics
1

Hi,

I want to use Phoenix for performing Average Bioequivalence (ABE) of a 2x2x2 design using the GLM model proposed by FDA in their guidance.

While I can use subject as a random effect, in Phoenix and perform ABE, this would become mixed effect model.

Can you please let me know how to use subject nested in sequence as a fixed effect while using phoenix.

2

Delete Subject(Seq) from random tab.

And add it to the end of Fixed Effects.

Remember it is case sensitive.

random.jpg
random.jpg548×223 20.9 KB
fixed.jpg
fixed.jpg530×225 21.9 KB

3

Phoenix uses the same algorithm with or without random effects.

  • it is like PROC MIXED, whether there are random effects or not.

Simon.

edit: actually to clarify if all effects are fixed then GLM is feasible
​
For the case when subject only is random and the design is not replicated it is still GLM, you may read a more detailed discussion on this thread https://forum.bebac.at/mix_entry.php?id=19512#p19524