Talk:False positive rate

Merge
This page doesnt seem to add anything over Type I and type II errors. I don't think having a separate page for this subject is helpful Purple Post-its  (talk) 14:41, 12 July 2011 (UTC)

I agree with removing this page or editing it substantially. As it stands, I think the definition on this page is simply wrong, or at a minimum is super nonstandard. The numerator is not part of the numerator in this definition, so it makes no sense. HarveyMotulsky (talk) 22:52, 3 June 2017 (UTC)

Aug 2012 expansion
Hello everyone, as part of my "selective inference" MSC coursework I have just finished expanding this term. Regarding the merge proposal above please see the new section "difference between type I error rate and false positive rate".

Miron.Avidan (talk) 04 Aug 2012

A different definition of false positive rate
A more obvious definition of false poaitive rate (or false discovery rate) is (in the notation used here) is TP/(TP +FP), That is the probability that a positive test is a false positive. It has been used in this sense by Colquhoun 2014 and 2017. This article needs to be rewritten, I think.