Estimating the Proportion of True Null Hypotheses, . . . (2003)
BibTeX
@MISC{Ferkingstad03estimatingthe,
author = {Egil Ferkingstad and Mette Langaas and Bo Lindqvist},
title = {Estimating the Proportion of True Null Hypotheses, . . .},
year = {2003}
}
OpenURL
Abstract
nd the use of kernel density estimation with a choice of smoothing parameter especially tailored to estimate ### . The estimators are derived under the assumption of independent p-values, and evaluated on simulated data with different degree of dependence. A discussion of the issue of modelling dependencies, with special emphasis on DNA microarray data analysis is presented. Finally, the estimators are applied to real data from DNA microarray experiments. CONTENTS i Contents 1 Introduction 1 2 Preliminaries, notation and model assumptions 4 2.1 Single hypothesis testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2 Multiple hypothesis testing framework and model assumptions . . . . . . . . . . . . 4 2.3 Mixture model and identifiability . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.4 Multiple testing error rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 of true null hypotheses? 7 3.1 The false discovery rate (FDR) . .







