Hi, Does Phoenix WNL output the 95% confidence upper bound for a replicate BE study or I have to extract the SwR and SE values from the anova tables to perform the computation? Thanks John
Jack - am I being stupid or are you looking for this; 
I just used the Bioequivalence object in the Toolbox options. Simon.
Hi Simon, I guess John is referring to RSABE according to FDA’s progesterone guidance. It is possible to set up a workflow (see Bioequivalence and Bioavailability Forum • SAS and missing values ) this post in BEBA-forum). You have to get the values of the inverse Chi ²-distribution from somewhere else (not accessible in Phoenix’ transformation functions). RSABE is demonstrated if sWR>0, the upper 95% CL (named “critbound” in the guidance) <0, and the PE is within 0.8000 and 1.2500. @John: Depending on your skills and experience it will take you 30 min to 2 h to set up the model in PHX. You can download EMA’s [http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500002963.pdf example datasets in Excel 2000 format http://bebac.at/downloads/Validation%20Replicate%20Design%20EMA.xls. As you can see from the post linked above it is possible to validate Phoenix against SAS. If you are not in a hurry I would suggest to wait until PHX6.3 is released. The new Data Wizard saves a lot of time and helps to keep the workflow clean.
Thanks Helmut, I ended up writing my own SAS code based on the FDA Progesterone guidance… John