Step-Down For Procedures For Large Numbers Of Hypotheses

Somerville (2004b) developed FDR step-down procedures which were particularly appropriate for cases where the number of false hypotheses was small. The test statistics were assumed to have a multivariate-t distribution with common correlation. MCV\u27s (minimum critical values) were chosen so that 8 unique critical values resulted. Tables were given for numbers of hypotheses m, ranging from 50 to 10,000, for p = 0., 0.5, and ν = 15, ∞. In this paper we extend the results, using MCV\u27s resulting in 31 critical values. Tables are given for the same values of m, for ρ = 0,0.1,0.5 and ν = 15, ∞. Interpolation rules are given for m, ρ and ν. Use of larger numbers of critical values increase both the power and the number of hypotheses falsely rejected. When the expected number of false hypotheses is small, use of the procedures of this paper results in a reduced number of false rejections with a negligible reduction in power. © Springer-Verlag Berlin Heidelberg 2006