Skip to main content

Table 2 Comparison of α and statistical power for iterated false positive rate. The values are the same as those given in Table 1 except the significance level (α) was iterated for the PE method until its power matched that of the MP method where α was held constant.

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

n1 = n2 = n ν0 αPE αMP FPPE FPMP PowerPE PowerMP
2 0 - 0.05 - 0.0505 - 0.2146
2 2 0.190 0.05 0.0514 0.0489 0.3374 0.3336
2 4 0.120 0.05 0.0483 0.0522 0.3783 0.3840
2 6 0.095 0.05 0.0496 0.0514 0.4302 0.4267
2 16 0.067 0.05 0.0491 0.0508 0.4794 0.4738
3 0 0.190 0.05 0.0507 0.0513 0.4576 0.4599
3 2 0.115 0.05 0.0531 0.0462 0.5324 0.5308
3 4 0.095 0.05 0.0475 0.0486 0.5611 0.5641
3 6 0.082 0.05 0.0536 0.0500 0.5994 0.6053
3 16 0.066 0.05 0.0511 0.0472 0.6531 0.6469