CODEWITHSUNDEEP
r
Denis
Denis

Reputation: 315

Adjust significance threshold (alpha) according to FDR (Benjamini & Hochberg)` method in R

I'm aware about p.adjust function in R and it works well for my needs. However, now i'd like to correct significance threshold (alpha) instead of p-values themselves according to FDR (Benjamini & Hochberg) method. For instance we have a ten of raw p-values:

0.0001,0.001,0.024,0.56,0.0077,0.55,0.0025,0.01,0.015,1

In case of Bonferroni it's very easy:

alpha_Bonferroni_corrected = 0.01/ number of tests (10 in our example)=0.001

But for FDR it will be a more tricky. Is there function in R for that?

Upvotes: 3

Views: 651

Answers (1)

paoloeusebi
paoloeusebi

Reputation: 1086

mutoss package seems to offer greater flexibility

 library(mutoss)
 alpha <- 0.01
 set.seed(1234)
 p <-c(runif(10, min=0, max=0.01), runif(10, min=0.9, max=1))
 result <- adaptiveBH(p, alpha)
 result

Upvotes: 1

Related Questions