Table 2 Simulation designs for type I error
From: Unsupervised gene set testing based on random matrix theory
Type I error design # | Covariance structure | g | p | n | σ 2 | ρ |
|---|---|---|---|---|---|---|
MVN-1 | Identity | 10 | 100 | 100 | 1 | 0 |
MVN-2 | Scaled Identity | 10 | 100 | 100 | 2 | 0 |
MVN-3 | Compound symmetry | 10 | 100 | 20 | 1 | 0.1 |
MVN-4 | Compound symmetry | 10 | 100 | 50 | 1 | 0.1 |
MVN-5 | Compound symmetry | 10 | 100 | 100 | 1 | 0.1 |
MVN-6 | Compound symmetry | 10 | 100 | 100 | 2 | 0.1 |
MVN-7 | Compound symmetry | 10 | 100 | 100 | 1 | 0.2 |
Binomial-1 | Compound symmetry | 10 | 100 | 100 | 0.375 | 0.1 |