Manual 3-step hits limit of nominal c... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Jon Heron posted on Tuesday, June 23, 2015 - 8:55 am
Hi Bengt/Linda

I estimated a parallel mixture model (C with C) and wished to related this classification with a distal binary outcome using the three-step method (Vermunt's one).

To do this I re-parameterised my 4-class by 5-class model as a 20-class model with various constraints, I then exported the modal-class assignment from this model and attempted to set up a manual 3-step model using the 19-by-20 matrix of logit constraints.

Once I had this all programmed I was met with the message that 10 categories is the limit for nominal variables.

Any tricks I might use to get round this?


many thanks, Jon
 Tihomir Asparouhov posted on Thursday, June 25, 2015 - 8:32 am
Jon

Is it not possible to use the method is Section 4 in
http://statmodel.com/download/webnotes/webnote15.pdf
 Jon Heron posted on Thursday, July 02, 2015 - 2:15 am
thanks Tihomir, will try that
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: