1

Based on what I have learned (and also this article: https://en.wikipedia.org/wiki/False_discovery_rate as well as the original 1995 paper), BH procedure requires specifies a FDR-value threshold. However, I have seen reported raw "adjusted pvalues" that simply adjust pvalues using BH, without specifying the procedure a FDR threshld. How precisely is this done without specifying the FDR threshold? (Previously asked here: What's the formula for the Benjamini-Hochberg adjusted p-value? but with no answer)

Josh
  • 471
  • 4
  • 15
  • I think you will still need to specify FDR threshold and get the q-value. The `qvalue` R-bioconductor package can do that for you. – SmallChess May 12 '17 at 16:45
  • http://genomine.org/papers/Storey_FDR_2011.pdf has the formula for the q-value. – SmallChess May 12 '17 at 16:45
  • The p.adjust package in R adjusts the pvalues without any FDR threshold supplied. – Josh May 12 '17 at 17:23
  • I just answered the original question (would it be useful to replicate it as an answer to this question?): https://stats.stackexchange.com/a/349686/112049 – Aditya Jun 05 '18 at 03:30

0 Answers0