I've been reading a lot about A/B testing and I understand the basic frequentist methods. I'm having a hard time with the Bayesian approach and was wondering if i could get an explanation of exactly how the Bayesian testing works and why it is favorable. I'm looking for a mathematical explanation.
4
-
$\begingroup$ An A/B test is a particular kind of experimental design but more than one kind of thing can be the subject of that sort of design. Are you just looking at a two sample comparison of count proportions (e.g. CTR) or a comparison of means (such as revenue, for example)? Or both? Or more possibilities than just those two comparisons? $\endgroup$– Glen_bCommented Oct 16 at 2:58
-
$\begingroup$ "why it is favorable" - are you sure it is? stats.stackexchange.com/questions/632935/… $\endgroup$– Christian HennigCommented Oct 16 at 9:21
-
$\begingroup$ @Glen_b both and more possibilities $\endgroup$– jenCommented Oct 16 at 17:35
-
$\begingroup$ Which ones did you want? Coverage of every plausible thing that could be subject to that design could be very long indeed. $\endgroup$– Glen_bCommented Oct 17 at 0:06
Add a comment
|