Bayesian Analysis of Clustered Regression Models for Mixed Discrete and Continuous Outcomes – An Application to Depressive Disorder Data

Abstract

A computationally practical form of Bayesian analysis for multiple response variables has been applied to data on depressive disorder. The model, considered here, allows for arbitrary covariate effects and also cluster effects through some known link functions. It is quite flexible and is applicable to any clustered data involving polytomous and continuous variables.

Keywords

Latent variables Gibbs sampler Shortened Rapid Eye Movement Hamilton Depressive Disorder Teratogenesis.

Get full access to this article

View all access options for this article.