mortonjt
commented
7 years ago Would also be really cool if we could take advantage of the existing work done with multivariate linear mixed effects models
https://www.jstatsoft.org/article/view/v033i02/v33i02.pdf http://sci2s.ugr.es/keel/pdf/specific/articulo/schafer_yucel02.pdf http://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/mixed-multivariate-responses.pdf
@serenejiang do you have any thoughts on this?
Right now, the main statistic being reported by statsmodels is the loglikelihood statistic
It would be great if we could support other meaningful model statistics (i.e. R^2)