Skip to main content
4 events
when toggle format what by license comment
Oct 2, 2015 at 8:50 vote accept en1
Jul 31, 2015 at 14:35 comment added en1 I understand that in the case of testing you don't have the same null, though I'm not sure if it makes any sense to say that a null is better than another null. But what about the case of prediction? Completely different story, still parametric vs. nonparametric dilemma.
Jul 31, 2015 at 13:58 history edited StasK CC BY-SA 3.0
deleted 2 characters in body
Jul 30, 2015 at 15:31 history answered StasK CC BY-SA 3.0