Williams design 3-way [Design Issues]

posted by vezz – Erba (CO), Italy, 2021-06-11 11:11 (1186 d 21:53 ago) – Posting: # 22406
Views: 11,186

Hi Helmut,

in your example you are assuming a model with no period effect, while I am considering the usual model for the analysis of BE studies, including the period effect.

With the period effect included in the model, the estimates of treatment effect are unbiased also using Brus' design.

Kind regards,

Stefano

Complete thread:

UA Flag
Activity
 Admin contact
23,221 posts in 4,877 threads, 1,656 registered users;
35 visitors (1 registered, 34 guests [including 8 identified bots]).
Forum time: 09:04 CEST (Europe/Vienna)

Explanations exist; they have existed for all time;
there is always an easy solution to every human problem –
neat, plausible and wrong.    H. L. Mencken

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5