A Bayesian paradox of confirmation

What is it about?

Shows that high posterior probabilities of equivalence are highly dependent on assumptions.

Featured Image

Why is it important?

A simple model is used to investigate the relevance of ‘competence’ to active control equivalence studies (ACES). It is shown that to the extent that such trials are successful the results of such trials must raise doubts regarding their competence. ACES are thus more problematic than classical clinical trials and the problems with such studies cannot be solved simply by exchanging the usual roles of null and alternative hypotheses.