4 ms·
Not necessarily. You can present your Bayesian approaches in such a way that it's almost independent of the prior. Your output will be 'this experiment should
by eru 3y ago
Not necessarily.
You can present your Bayesian approaches in such a way that it's almost independent of the prior. Your output will be 'this experiment should shift your odds-ratio by so-and-so-many logits in this or that direction' instead of an absolute probability.