Bioequivalence and Bioavailability Forum • Power of Proc Power

Power of Proc Power [Power / Sample Size]

posted by d_labes – Berlin, Germany, 2019-03-20 19:37 (1862 d 21:06 ago) – Posting: # 20058
Views: 6,295

Dear sury,

Additionally to what Helmut has said:

❝ we have not yet finalized the SOP, and moreover we use the SAS Software procedure "PROC POWER" for the estimation of the sample size. But however we wanted it to be more accurate sample size for that we want to include the deletti tables.

PROC POWER doesn't allow a power calculation or a sample size estimation out of the box for study designs used mainly in bioequivalence studies, beside other f.i. for the standard 2x2 crossover design! Except if you assume no period effects present. See here.
But this assumption isn't appropriate in many cases.

You have to roll your own piece of software. See f.i. this paper.
And if you do so: You have to validate.

—
Regards,

Detlew

Complete thread:

Deletti tables for sample size sury 2019-03-19 12:22
- Of historic(al) interest only Helmut 2019-03-19 13:27
  - Of historic(al) interest only sury 2019-03-20 05:04
    - Software validation – again and again Helmut 2019-03-20 12:12
      - Software validation – again and again nobody 2019-03-20 16:20
        
        Fiaker und Apfelstrudel Helmut 2019-03-20 16:35
        
        Fiaker und Apfelstrudel nobody 2019-03-20 17:34
        
        Fiaker und Apfelstrudel Helmut 2019-03-20 17:54
        
        Fiaker und Apfelstrudel d_labes 2019-03-20 18:39
        Fiaker und Apfelstrudel ElMaestro 2019-03-20 22:09
      - White-box validation mittyri 2019-03-20 21:09
        
        White-box validation Helmut 2019-03-21 00:19
    - Power of Proc Powerd_labes 2019-03-20 18:37