Latent-basis GCM with individually-va... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Patrick Coulombe posted on Tuesday, April 11, 2017 - 11:25 am

I came across a paper that shows a method to estimate a shape-factor growth curve model (or latent-basis or fully-latent GCM or whatever it's called, where the loadings are estimated from the data) with individually-varying times of observation.

Here's the part where the author describes the matrix of factor loadings:

Matrix of factor loadings with individually-varying times of observation

And here's the corresponding Mplus syntax:

Corresponding Mplus syntax

I was surprised to see the time variables (with time values that vary from person to person) in the constraint section, and I was also surprised that this method to accommodate individually-varying times of observation does not use the TSCORES=... option in conjunction with "AT" in the MODEL command. So I wanted to check in: is this indeed doing what the paper says it's doing (i.e., using different factor-loading matrices for each person)?
 Bengt O. Muthen posted on Thursday, April 13, 2017 - 4:40 pm
The best source for this is the new book Growth Modeling by Grimm, Ram, Estabrook.
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