Regression models for correlated binary data with random effects assuming a mixture of normal distributions (Q1424628): Difference between revisions

A logistic regression model with random effects is considered for longitudinal binary data. It is assumed that the random effects distribution is a mixture of normal distributions with different unknown means and variances. A Bayesian approach with Markov chain Monte-Carlo (MCMC) is described for the parameter estimation. An example of biological data analysis is presented.