Hi everyone, I have a set of latent class models fit using Latent Gold using different sets of both nominal and continous indicator variables and different numbers of classes (clusters). Can BIC be used to compare and contrast model fitness to the data when both the number of classes and the set of indicator variables differ among models? If anyone has any good references on this subject, I would like to read them.
Thank you, Dan ---------------------------------------------- CLASS-L list. Instructions: http://www.classification-society.org/csna/lists.html#class-l
