Abstract

Genes and environmental components lead to the development of several complex diseases and the analyses of these factors play a major role in the prevention and/or control of the diseases in human population. The analysis is, however, not easy because of the fact that genes and environmental factors interplay while causing diseases. If this interaction is ignored at the time of estimation of the contributions of these factors, it may provide incorrect estimates of the proportions of the disease caused by these factors. The case-control study, where sampling is conditional on the presence or absence of the disease, is a powerful epidemiological tool for studying such problems and the Bayesian framework offers increased level of flexibility for the possible modelling. A well known and user-friendly approach to analyze such data is multinomial-Dirichlet modeling assumption. The present paper analyzes a retrospective data on ovarian cancer based on the assumption of multinomial-Dirichlet model. Empirical Bayes method is used to select the prior hyperparameters. Results are found to be logical and appealing.

Author: Akanksha Gupta and S. K. Upadhyay

Received on: November, 2013

Accepted on: April, 2014