How to free residual variances? PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Charlotte Vrijen posted on Tuesday, February 18, 2020 - 9:21 am
Hello,

I'm fitting a growth mixture model on continuous indicators from four time points but have trouble arriving at a sounds solution. I keep getting non-positive latent variable covariance matrices and someone suggested to check residual variances. Could you tell me how I can free the residual variances of the indicator items across classes and/or across time in the growth mixture model?

Thanks in advance!

Charlotte
 Bengt O. Muthen posted on Tuesday, February 18, 2020 - 5:22 pm
We need to see your full output to give advice on this - send to Support along with your license number.
 Charlotte Vrijen posted on Tuesday, February 18, 2020 - 11:34 pm
Thanks, but would it also be possible to provide the syntax I can use to free the residual variances of the indicator items across classes and/or across time? Thanks in advance!
 Bengt O. Muthen posted on Thursday, February 20, 2020 - 4:35 pm
I don't think letting residual variances vary across classes will solve this problem. They are already free across time as the default.
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: