Some methods to adjust p-values for multiple comparisons in Julia. The various methods can be called using the function padjust. padjust takes an array of p-values and a second method parameter and returns an array of adjusted p-values. Please refer to the documentation of the corresponding function in R or to Wikipedia (FWER, FDR), if you want to know more on this topic.

Current stable version is 2.0.0.

In case you find any bugs please post an issue here or send a pull request. Make sure you write a test for your contribution.

# get the latest stable version
Pkg.add("PValueAdjust")

# get the latest development version
Pkg.clone("git:https://github.com/dirkschumacher/PValueAdjust.jl.git")

julia > pValues = [0.05, 0.03, 0.01, 0.5]
julia > padjust(pValues, Bonferroni)
4-element Array{Float64,1}:
 0.2 
 0.12
 0.04
 1.0

julia > pValues = [0.05, 0.03, 0.01, 0.5]
julia > padjust(pValues, Hochberg)
4-element Array{Float64,1}:
 0.1 
 0.09
 0.04
 0.5

Also known as the Holm–Bonferroni method.

julia > pValues = [0.05, 0.03, 0.01, 0.5]
julia > padjust(pValues, Holm)
4-element Array{Float64,1}:
 0.1 
 0.09
 0.04
 0.5

julia > pValues = [0.05, 0.03, 0.01, 0.5]
julia > padjust(pValues, BenjaminiHochberg)
4-element Array{Float64,1}:
 0.0666667
 0.06     
 0.04     
 0.5

julia > pValues = [0.05, 0.03, 0.01, 0.5]
julia > padjust(pValues, BenjaminiYekutieli)
4-element Array{Float64,1}:
 0.138889 
 0.125    
 0.0833333
 1.0

This package uses Semantic Versioning 2.0.

Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B 57, 289–300.

Benjamini, Y., and Yekutieli, D. (2001). The control of the false discovery rate in multiple testing under dependency. Annals of Statistics 29, 1165–1188.

Hochberg, Y. (1988). A sharper Bonferroni procedure for multiple tests of significance. Biometrika 75, 800–803.

Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics 6, 65–70.