Bioequivalence and Bioavailability Forum

Hi all,

I am trying to figure out if we can say anything general about the sigma (let's call it s) we get from a 222BE study versus the individual treatment within-sigmas we get from a 224-study (s_T and s_R).

My guess is that the expected value of s will fall in between s_T and s_R regardless of the magnitude of the fixed treatment effects, but I have no real idea if that holds true. Another way of saying it is that in a simulation experiment s will be in between s_T and s_R for infinitely big sample sizes.

I am probably wrong, though? I am saying this not because I have a mathematical proof but rather because I generally tend to screw up badly when I try to use my sad excuse for a brain.

Any insight here? Many thanks in advance.