Premium
A Note on Controlling the Number of False Positives
Author(s) -
Korn Edward L.,
Freidlin Boris
Publication year - 2008
Publication title -
biometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.298
H-Index - 130
eISSN - 1541-0420
pISSN - 0006-341X
DOI - 10.1111/j.1541-0420.2007.00881.x
Subject(s) - bonferroni correction , permutation (music) , false positive paradox , multiple comparisons problem , statistics , mathematics , multivariate statistics , type i and type ii errors , physics , acoustics
Summary : Lehmann and Romano (2005, Annals of Statistics 33, 1138–1154) discuss a Bonferroni‐type procedure that bounds the probability that the number of false positives is larger than a specified number. We note that this procedure will have poor power as compared to a multivariate permutation test type procedure when the experimental design accommodates a permutation test. An example is given involving gene expression microarray data of breast cancer tumors.