Table 3 Comparison of α and statistical power for iterated false positive rate and variable variance. The values are the same as those given in Table 1 and Table 2 except the variance for each sample was drawn from a uniform distribution between 0.05 and 1 during the simulation. The significance level (α) for the PE method was iterated so that the observed false positive rate matched that of the MP method where α was held constant.

From: A two-sample Bayesian t-test for microarray data