Fixing classes PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Corinna posted on Thursday, September 10, 2015 - 6:39 am
Dear Linda & Bengt,
I am analyzing a GMM or LCGA respectively and want to free variances of one specific class with significant and high variances but want to fix the variances of the other classes to zero because they are insignificant. However, I do not know how I refer to the specific class in which I want to fix or free the variances. The classes change from analysis to analysis and so I do not know which is the class I want to fix... I heard that it is possible to work the fixation of growth factors to specific values in order to do that?
I would be very thankful if you could give me the syntax or tell me how to do this. Thank you very much,
best regards
 Bengt O. Muthen posted on Thursday, September 10, 2015 - 9:05 pm
You can give good starting values for the growth factor means and use Starts=0 and thereby get the right class to have zero variance.

Or, you can leave things as they are and let the Mplus search over the likelihood to find the class that most benefits from a zero variance. That is, you don't give starting values; you just fix variance to zero in a certain class. The the STARTS will do the search.
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: