A Bayesian Approach for Estimating Twin Similarity with Respect to a Binary Trait Under the Beta-Binomial Model

Ahmed, Marwa; Shoukri, Mohamed

doi:10.21608/esju.2011.314308

A Bayesian Approach for Estimating Twin Similarity with Respect to a Binary Trait Under the Beta-Binomial Model

Document Type : Original Article

Authors

Marwa Ahmed ¹
Mohamed Shoukri ²

¹ Faculty of Graduate Studies and Statistical Research, Cairo University

² Department of Biostatistics, Epidemiology and Scientific Computing, King Faisal Sepecialist Hospital and Research Center, Riyadh, Saudi Arabia.

10.21608/esju.2011.314308

Abstract

In twin studies there is a need to compare a sample of monozygotic to another independent sample of dizygotic twins with respect to the presence or absence of a specific disease or condition. A significant difference between the intracluster correlation coefficients between the two independent samples indicates the presence of a genetic component in the disease etiology, and is an estimate of twin similarity. In this paper, Bayesian statistical methodology is presented for the estimation of twin similarity with respect to a dichotomous trait. The parameter of interest is the difference between the intracluster correlation coefficients (ICC) of the two zygosity groups. Under the Beta- Binomial model, point and interval estimators for this difference, 4, are obtained.

Keywords

Main Subjects