Doctoral Dissertations
Keywords and Phrases
P-value, gene expression, high dimension, multiple testing, mixture model, microarray experiment
Abstract
"High dimensional experiments (HDEs) involving thousands of simultaneous hypothesis tests have become more prevalent in recent years. Several statistical methods that have been proposed for analyzing data from such experiments have used the distribution of P-values that result from these tests. This dissertation brings together some past research on the distribution of a P-value as a random variable and places it in the context of HDEs. One scenario illustrates a non-unifonn null distribution for a P-value and the effect on interpretation of results from HD Es, and another considers the effect of correlated P-values on a key result that is often of interest in HD Es. The distribution of a P-value under the alternative hypothesis is explored when the effect size is a fixed value or is a random variable under some distribution. The suitability of a beta distribution for modeling the true distribution is discussed. This disse1iation reveals some insights into the distribution of P-values from multiple tests, some challenges that can arise when modeling it, and some suggested approaches to the challenges"--Abstract, page iii.
Advisor(s)
Gadbury, Gary L.
Committee Member(s)
Le, Vy Khoi
Samaranayake, V. A.
Westenberg, David J.
Wen, Xuerong
Department(s)
Mathematics and Statistics
Degree Name
Ph. D. in Mathematics
Sponsor(s)
National Science Foundation (U.S.)
Publisher
University of Missouri--Rolla
Publication Date
Spring 2007
Pagination
x, 88 pages
Note about bibliography
Includes bibliographical references (pages 84-87).
Rights
© 2007 Xiaojun Hu, All rights reserved.
Document Type
Dissertation - Restricted Access
File Type
text
Language
English
Subject Headings
Gene expressionMultiple comparisons (Statistics)
Thesis Number
T 9182
Print OCLC #
173259973
Recommended Citation
Hu, Xiaojun, "Distributional Aspects of P-value and Their use in Multiple Testing Situations" (2007). Doctoral Dissertations. 1722.
https://scholarsmine.mst.edu/doctoral_dissertations/1722
Share My Dissertation If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.
