A Bayesian Semi-Parametric Approach to Variable Selection

Identifier

1330

Author

William G. Terry

Date

2015

Document Type

Thesis

Degree Name

Master of Science

Major

Mathematical Sciences

Concentration

Statistics

Committee Chair

Hongmei Zhang

Committee Member

Dale Bowman

Committee Member

Ebenezer George

Committee Member

Lih Deng

Abstract

Using semi-parametric models with Gaussian kernels, a variable selection process is proposed. Through simulations, the proposed method, which can select either continuous or discrete variables which may have complex interactions, is evaluated. The method is employed in a real data application to identify important CpG sites and single nucleotide polymorphisms (SNPs) in a set of genes such that DNA methylation of those sites and SNPs may have joint effects on allergic sensitization.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Recommended Citation

Terry, William G., "A Bayesian Semi-Parametric Approach to Variable Selection" (2015). Electronic Theses and Dissertations. 1120.
https://digitalcommons.memphis.edu/etd/1120