Reordering latent classes, output pro... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Aniruddha Das.  posted on Sunday, December 30, 2007 - 2:22 pm
Dear Drs. Muthen,
I'm new to Mplus, and had a quick question -- how precisely do I use the DEFINE command to reorder the latent classes in a mixture model, to regress on covariates? I'm a little confused about the syntax, and any help would be appreciated.
I've also experienced a problem when, *after* convergence [the TECH8 screen shows the best log likelihood is replicated] it takes hours for the output to appear. I'm using version 4.21. The particular model is a factor mixture model, where only the means of the factors (4) are varying across the classes (5). The data is stratified and clustered, if that's an issue. Thanks again,
Bobby Das.
 Linda K. Muthen posted on Sunday, December 30, 2007 - 4:20 pm
The DEFINE command is used to create new variables or transform existing variables. It is not used to reorder latent classes. If you want to order the classes in a specific way, you use the parameter estimates as starting values.

A factor mixture model can take considerable time depending on several factors. Please send your input, data, output, and license number to support@statmodel.com so we can see what about your model is taking time.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: