http://www.stat.columbia.edu/~cook/movabletype/archives/2008/03/multivariate_mu.html Dave Judkins wrote:... en-us <p>Dave Judkins wrote:</p> http://www.stat.columbia.edu/~cook/movabletype/archives/2008/03/multivariate_mu.html <p>"This matters because we have a mixed set of binary and continuous outcomes. Trying to fit them into a unified multivariate multi-level model is a pain. I would like to make sure that there is really some benefit to be had."</p> <p>Yes, multivariate models can be a pain to fit, especially if you have many variables. Gain? You will gain correlations and additionally the information on one variable is used also in estimation of parameters for other variable and vice versa. Take a look at http://dx.doi.org/10.1051/gse:2003002 for a general approach on modelling "Multivariate {B}ayesian analysis of {G}aussian, right censored {G}aussian, ordered categorical and binary traits using {G}ibbs sampling"</p> http://www.stat.columbia.edu/~cook/movabletype/archives/001600.html#544056