Bivariate high-level exceedance and the Chen–Stein theorem in genomics multiple hypothesis testing perspectives

In genomic studies, generally the genes are neither independent nor marginally identically distributed, though in microarray studies and DNA/RNA SNP models, often they are assumed to be independent and identically distributed. A version of the Chen–Stein theorem on Poisson approximation for dependen...

Full description

Saved in:
Bibliographic Details
Published inStatistics & probability letters Vol. 83; no. 7; pp. 1725 - 1730
Main Authors Sen, Pranab K., Kang, Moonsu
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.07.2013
Subjects
Bivariate extreme
FDR
FWER
HDLSS
Microarray data model
Microarray data model
FDR
Bivariate extreme
HDLSS
FWER
Online AccessGet full text

Cover

More Information
Summary:In genomic studies, generally the genes are neither independent nor marginally identically distributed, though in microarray studies and DNA/RNA SNP models, often they are assumed to be independent and identically distributed. A version of the Chen–Stein theorem on Poisson approximation for dependent binary variables has been adopted for a mathematical justification of this approach in a general genomic setup.
ISSN:0167-7152
1879-2103
DOI:10.1016/j.spl.2013.03.019