Premium
Bivariate random effect model using skew‐normal distribution with application to HIV‐RNA
Author(s) -
Ghosh Pulak,
Branco Marcia D.,
Chakraborty Hrishikesh
Publication year - 2006
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.2667
Subject(s) - skewness , bivariate analysis , random effects model , normality , multivariate normal distribution , econometrics , data set , multivariate statistics , bayesian probability , skew , statistics , computer science , mixed model , skew normal distribution , mathematics , medicine , telecommunications , meta analysis
Correlated data arise in a longitudinal studies from epidemiological and clinical research. Random effects models are commonly used to model correlated data. Mostly in the longitudinal data setting we assume that the random effects and within subject errors are normally distributed. However, the normality assumption may not always give robust results, particularly if the data exhibit skewness. In this paper, we develop a Bayesian approach to bivariate mixed model and relax the normality assumption by using a multivariate skew‐normal distribution. Specifically, we compare various potential models and illustrate the procedure using a real data set from HIV study. Copyright © 2006 John Wiley & Sons, Ltd.