Figure 2

False discovery rate with different sample sizes. Percentage (%) of FDR for different sample sizes. The family of tests was 10,000 one-sample t tests with 5% of them coming from a N (0.36, 1). Values are averages through 1,000 replicates.