On the Use of a Logistic Regression Model in the Gene-Environment Problem: A Bayesian Approach

SYNOPTIC ABSTRACT

The article provides the Bayes analysis of case-control data related to growth of cancerous cells in a reproductive organ of women. It is claimed that such cancerous growth may occur because of genes and environmental effects, the latter may be categorized as some personal practices, food and drinking habits, etc. The present work considers the logistic regression model with a number of genetic and environmental covariates to analyze such a data set. The ultimate aim of the study is to see if there is any association between gene and environmental components in the development of the disease. Some comments are also made to simplify the proposed logistic regression model by looking at appropriate variable selections using an important tool of Bayesian paradigm. Numerical results based on a real data set are provided for illustration purposes.

KEYWORDS AND PHRASES:

• Case-control study
• retrospective analysis
• gene-environment interaction
• logistic regression

Acknowledgements

The authors express their thankfulness to the Editor and the referees for their valuable comments and suggestions that improved an earlier version of the article.