Bioequivalence and Bioavailability Forum

Dear dlabes,

❝ Is it because an Old pirate always knows from where the wind blows?

❝ Or had you also such a question in your career?

If you put an old sock into someones mouth, he or she won't say much. In this case I just try to imagine from the wording you gave above what type of sock would fit.

❝ Is the Proc MIXED result with REML estimates to complicated to understand?

Yes, honestly, the process goes via maximizing the likelihood of a totally osbcure multi-matrix expression. So if the process if complicated, why shoudn't the result be? Bear in mind, the assessor might not even be a statistician, but a lowly paid slave (biologist, pharmacist, etc) who is trying to impress his or her boss by asking questions that indicate a profound understanding of everything.

❝ Or why do they not trust them?

❝ Or is it because the French do not love Americans?

Airbus vs. Boeing. And they started to make their own wine over there, so less Bourgogne is being exported. As a counter-measure the French have stopped paying for their meals at McDonalds*.

❝ BTW: Why did you suggest all the summary statistics above? The only relevant is the variance I think, because it is well known that T-T or R-R have the expectation 2*sigma²_within for test or reference.

You gave the wording from the assessor's question above. (S)he did not mention variance, but variability, so even though I agree that sigma or variance is most relevant here this might not itself cover variability. It is a matter of making the sock large enough.

EM.