LCGA/GMM with time-varying covariates PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Jennifer Rose posted on Thursday, March 06, 2008 - 5:36 pm
Is it possible to test a LCGA model or GMM with a time-varying covariate?

Thanks.
 Linda K. Muthen posted on Thursday, March 06, 2008 - 9:01 pm
Yes. You can see how to specify this in Example 6.12.
 Jennifer Rose posted on Friday, March 07, 2008 - 12:34 pm
Okay. I did use example 6.12 to set up the analysis, but I got an error message that I need to specify type=random if I want to include random effects for the time-varying covariate. I had it specified as type=mixture. If I change it to type=random, then I get an error message that I have to use type=mixture if I want to specify a latent class variable. Any suggestions for what I can do differently?

Thanks.
 Linda K. Muthen posted on Friday, March 07, 2008 - 12:40 pm
How about TYPE= MIXTURE RANDOM?
 Jennifer Rose posted on Friday, March 07, 2008 - 2:21 pm
I tried that and got the following error:

*** INTERNAL ERROR
Internal Error Code: MDP1044
An internal error has occurred. Please contact us about the error,
providing both the input and data files if possible.

I assumed that I was asking for something just wasn't possible. Should I send you the files associated with that error?
 Linda K. Muthen posted on Friday, March 07, 2008 - 2:50 pm
Please send the input, data, output, and your license number to support@statmodel.com.
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: