Transcription of 1 Why is multiple testing a problem?
{{id}} {{{paragraph}}}
Spring 2008 - Stat C141/ Bioeng C141 - Statistics for BioinformaticsCourse Website: Website: Contact Info:Megan Hours: 342 Evans M 10-11, Th 3-4, and by appointment1 Why is multiple testing a problem? Say you have a set of hypotheses that you wish to test simultaneously. The first idea thatmight come to mind is to test each hypothesis separately, using some level of significance .At first blush, this doesn t seem like a bad idea. However, consider a case where you have20 hypotheses to test, and a significance level of What s the probability of observingat least one significant result just due to chance?P(at least one significant result) = 1 P(no significant results)= 1 (1 )20 , with 20 tests being considered, we have a 64% chance of observing at least one sig-nificant result, even if all of the tests are actually not significant.
controlling the false discovery rate (FDR). This is de ned as the proportion of false positives among all signi cant results. The FDR works by estimating some rejection region so that, on average, FDR < . 4 The positive False Discovery Rate The positive false discovery rate (pFDR) is a bit of a wrinkle on the FDR. Here, you try to
Domain:
Source:
Link to this page:
Please notify us if you found a problem with this document:
{{id}} {{{paragraph}}}