Dear R-users, I would like to use the package flexmix to fit latent classes to a regression model. My data are repeated measurements of bernouilli variables so I can use the binomial family link to the glm function. The design is not balanced, meaning that for some individuals in my data set I have 10 measurements or more, for others I only have 5 or even less. My question is the following. Can flexmix handle this unbalancedness in the dataset (I looked in the jstatsoft.org paper "Flexmix: A General Framework for Finite Mixture Models and Latent Class Regression in R" and it only mentioned a dataset with balanced repeated measurements - 50 persons with 4 measurements each). When I look at the formula for the loglikelihood of the data, it seems to overweigh individuals with more measurements in an unbalanced design, am I correct in this and are there ways to correct for this? Thanks for the help, Jan Jan Wijffels University Center for Statistics W. de Croylaan 54 3001 Heverlee Belgium tel: +32 (0)16 322784 fax: +32 (0)16 322831 <http://www.kuleuven.be/ucs> http://www.kuleuven.be/ucs Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm [[alternative HTML version deleted]]