All ETDs from UAB

Advisory Committee Chair

Xiangqin Cui

Advisory Committee Members

Immaculada Aban

Degui Zhi

Lisa Schwiebert

Victor Darley-Usmar

Document Type

Dissertation

Date of Award

2013

Degree Name by School

Doctor of Philosophy (PhD) School of Public Health

Abstract

In this dissertation we have adapted two multi-group equivalence tests to be performed on high-dimensional data. The F and Range test for multi-group equivalence were applied to the public microarray dataset GSE11291 to detect equivalently expressed genes. They were also evaluated in terms of type I error and power using single gene simulation and a high-dimension simulation. The F test has higher power than the Range for the same simulated data and parameter settings, in the single gene simulations. The power of the two tests is similar in the high-dimension simulation. The multi-group equivalence tests that were applied to the microarray dataset were used to create the R package EquivMulti. The package offers multi-group testing utilizing the F and Range tests of equivalence with the adaptations discussed in Finally, in a clinical trial setting we use simulation study and real data analysis to analyze the behavior of two existing multivariate equivalence tests, Hotellings T2 testand the intersection union test (IUT). We also propose an alternative multivariate equivalence test, Max Test, which we evaluate along with the others using simulation study and real data analysis.

Included in

Public Health Commons

Share

COinS