Per-comparison error rate

In statistics, per-comparison error rate (PCER) is the probability of a result in the absence of any formal multiple hypothesis testing correction.[1] Typically, when considering a result under many hypotheses, some tests will give false positives; many statisticians make use of Bonferroni correction, false discovery rate, and other methods to determine the odds of a negative result appearing to be positive.

References

  1. ↑ Benjamini, Yoav; Hochberg, Yosef (1995). "Controlling the false discovery rate: a practical and powerful approach to multiple testing" (PDF). Journal of the Royal Statistical Society, Series B 57 (1): 289–300. MR 1325392.
This article is issued from Wikipedia - version of the Thursday, July 23, 2015. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.