MGA in parallel process model PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 JM posted on Monday, March 09, 2009 - 3:49 pm
Hi, I'm trying to run at MGA in my PPM with single indicators with non-normal data (using MLR estimator). My group codes are 0=uc and 1=qi, when I free the factor loadings to compare the two models as follows:

MODEL: level1 BY aydepsx1-aydepsx4@1;
trend1 BY aydepsx1@0 aydepsx2 Aydepsx3 aydepsx4@1;
[aydepsx1-aydepsx4@0];
[level1 trend1];
MODEL qi: level1 BY aydepsx1-aydepsx4@1;
trend1 BY aydepsx1@0 aydepsx2 aydepsx3 aydepsx4@1;
[aydepsx1-aydepsx4@0];
[level1 trend1];

The analysis fails to run (non convergence) only when I use MODEL qi as compared to when I run MODEL uc. Does the second model command have to be the group assigned as =1 vs. =0 dummy code?
 Linda K. Muthen posted on Monday, March 09, 2009 - 11:57 pm
The group you choose should make no difference. Please send your input, data, output, and 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: