A Review of Optimal Procedures for Combining P-Values with a Proposal of a Bayesian Approach to Identify Auxiliary Covariates

Author

Tiffany Geistkemper

Date

2024

Document Type

Thesis

Degree Name

Master of Science

Department

Mathematical Sciences

Committee Chair

Ebenezer George

Committee Member

Baris Kopruluoglu

Committee Member

Majid Noroozi

Abstract

In this thesis, we review a meta-analytic procedure for evaluating tests from several studies based on the combination of P -values. We provide a summary of properties of several well-known methods that combine P-values of continuous and independent test statistics. We note that it is not uncommon that, in the process of performing statistical tests, potentially important covariates are sometimes inadvertently not accounted for in the individual studies. We outline a procedure based on a Bayesian regression approach that introduces latent variables that may be used to account for and evaluate the significance of such covariates.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest.

Notes

Open Access

Recommended Citation

Geistkemper, Tiffany, "A Review of Optimal Procedures for Combining P-Values with a Proposal of a Bayesian Approach to Identify Auxiliary Covariates" (2024). Electronic Theses and Dissertations. 3579.
https://digitalcommons.memphis.edu/etd/3579