Bioequivalence and Bioavailability Forum • Anova table in Phoenix using the EMA model

mittyri
★★

Russia,
2015-11-16 07:27
(3443 d 22:39 ago)

Posting: # 15638
Views: 10,854

Anova table in Phoenix using the EMA model [Software]

Dear All,

Could you please guide me with the following question, may be it's arised due to weak understandig of ANOVA.
When Phoenix is building ANOVA model as default, the Sums of squares and the means are presented in the resulting table. But when I transfer 'subject in sequence' to the fixed parameters, ANOVA table is changing - no MS and SS, only nom_df and denom_df.
Why I cannot see the SSs?

—
Kind regards,
Mittyri

Helmut
★★★

Vienna, Austria,
2015-11-17 16:00
(3442 d 14:06 ago)

@ mittyri
Posting: # 15645
Views: 8,970

Get v6.4

Post reply

Hi Mittyri,

❝ When Phoenix is building ANOVA model as default, the Sums of squares and the means are presented in the resulting table. But when I transfer 'subject in sequence' to the fixed parameters, ANOVA table is changing - no MS and SS, only nom_df and denom_df.

It should be the other way ’round. In Phoenix WinNonlin 6.4:

[image]

Edit: Confirmed your observations in Phoenix WinNonlin 6.3. From the 6.4 Release Notes (p4, 08/2014):

For fixed effects models, an “old-fashioned ANOVA table” is now provided in LinMix/Bioequivalence output, that includes SS and MS terms (QC 9181): The LinMix/Bioequivalence objects now create additional Partial SS and Sequential SS output worksheets with SS and MS terms for models with all fixed effects.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Artem Gusev ☆ Russia, Moscow, 2017-04-26 19:03 (2916 d 12:03 ago) @ Helmut Posting: # 17268 Views: 6,293	Get v6.4 Post reply
	Dear Helmut! Can you attach screenshots with setup conditions for each type of BE, please? It might be helpfull for me. Gratefully, Artem. — Best Regards, Artem

Helmut
★★★

Vienna, Austria,
2017-04-26 21:16
(2916 d 09:49 ago)

@ Artem Gusev
Posting: # 17273
Views: 6,362

Get v6.4

Post reply

Hi Artem,

for the EMA (and agencies recommending its approach) all effects fixed.
In the Bioequivalence object > Fixed Effects > Model Specification:
sequence+subject(sequence)+period+treatment
Make sure that in Variance Structure both Random 1 and Repeated are empty.
For agencies recommending a mixed-effects model (FDA, Health Canada):
In the Bioequivalence object > Fixed Effects > Model Specification:
sequence+period+treatment
In Variance Structure Random 1: subject(sequence)

Out of the box Phoenix/WinNonlin suggests the mixed-effect model. Since it is annoying to change the specification if you have to deal mainly with the EMA’s model, there is an option since v6.4:
Edit > Preferences… > LinMixBioequivalence ☑ Default for 2x2 crossover set to all fixed effects

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Artem Gusev
☆

Russia, Moscow,
2017-04-27 11:12
(2915 d 19:53 ago)

(edited on 2017-04-27 11:32)
@ Helmut
Posting: # 17274
Views: 6,602

Get v6.4

Post reply

Dear Helmut!

Thank you for quick response.
I want to clarify one more point. I need to calculate additional parameter in ANOVA - Cohort factor.

Then in the case of a EMA Model Specification is:
sequence+subject(sequence)+period+treatment+cohort
Am I right?

And how Model Specification can be constructed for agencies recommending a mixed-effects model (FDA, Health Canada)?

Thank you!

Edit: This thread gets less software-related. I opened a new thread for answers. [Helmut]

—
Best Regards,
Artem