Test Multiple Comparisons Using Hommel's Method

Given a set of p-values and a test level, returns vector of test results for each hypothesis.

hommel.test(p, alpha=0.05)

Arguments

p

numeric vector of p-values

alpha

numeric value, desired significance level

Value

logical vector indicating whether each hypothesis is accepted

Details

This function implements the multiple testing procedure of Hommel (1988). Hommel's method is also implemented as an adjusted p-value method in the function p.adjust but the accept/reject approach used here is faster.

References

Hommel, G. (1988). A stagewise rejective multiple test procedure based on a modified Bonferroni test. Biometrika, 75, 383-386.

Shaffer, J. P. (1995). Multiple hypothesis testing. Annual Review of Psychology 46, 561-576. (An excellent review of the area.)

Author

Gordon Smyth

See also

p.adjust

Examples

p <- sort(runif(100))[1:10]
cbind(p,p.adjust(p,"hommel"),hommel.test(p))
#>                p             
#>  [1,] 0.01004905 0.05757171 1
#>  [2,] 0.01964032 0.05956709 1
#>  [3,] 0.02293579 0.05956709 1
#>  [4,] 0.03348343 0.05956709 1
#>  [5,] 0.03487595 0.05956709 1
#>  [6,] 0.03573694 0.05956709 1
#>  [7,] 0.04540150 0.05956709 1
#>  [8,] 0.04671066 0.05956709 1
#>  [9,] 0.04797642 0.05956709 1
#> [10,] 0.05956709 0.05956709 1