GMM Residuals PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Natalie D. Eggum posted on Friday, February 19, 2010 - 5:08 pm
I am running a growth mixture model and do not know how I should treat the residuals. It seems to me the residuals should be allowed to vary across latent class. Is this correct?

Within class, should I constrain them to be equal across time or allow them to be freely estimated across time?

Thank you!
 Linda K. Muthen posted on Saturday, February 20, 2010 - 3:49 pm
The Mplus defaults are that the residual variances are free across time and equal across classes. I would start with the defaults and look at plots to see if the equality across classes needs to relaxed for one or more classes.
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: