Mixture sem models with longitudinal ... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Raji Srinivasan posted on Wednesday, January 11, 2006 - 6:53 pm
I am new to mplus.
I have to implement mixture sem models using panel data (with repeated measures).

I assume that this is possible. Am I correct?

 Linda K. Muthen posted on Thursday, January 12, 2006 - 8:41 am
Mplus can estimate growth mixture models.
 Marco Pannacci posted on Tuesday, September 26, 2017 - 6:06 am
Dear Mplus support,
I estimated a longitudinal factor analysis where my latent variable SAD is measured with 3 observed variables (lung1, lung2, lung3) at 3 different timepoints (see the code below).

I would like to perform a mixture model where for each SAD1, SAD2, SAD3 I would like to link a categorical variable C1 (for SAD1), C2 and C3 (in order to obtain 3 clustering at 3 timepoints based on SAD latent variables).
How can I do it in the MODEL statement?

Many thanks
Marco Pannacci


SAD1 by lung11* lung21 lung31@1 (d1-d3);

SAD2 by lung12* lung22 lung32@1 (d1-d3);

SAD3 by lung13* lung23 lung33@1 (d1-d3);
 Bengt O. Muthen posted on Tuesday, September 26, 2017 - 6:06 pm
Take a look at UG ex 8.7. Although this refers to a growth model, your factor model is similar enough. The different classes can have different factor means if you add measurement invariance with respect to the intercepts of the factor indicators.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message