Interesting Bayesian decision analysis of whether the FDA is too conservative: alo.mit.edu/wp-content/upl.
Conversation
Replying to
Bayesianism feels so odd here because you still have to assign weighted costs to type I & II error! Like obfuscated god-playing.
Replying to
Agreed -- but I don't see how you can possibly avoid it.
1
Oh, I totally don’t either. It’s the obfuscation that feels weird to me!