Variance and covariance PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 lisa Car posted on Thursday, June 14, 2018 - 6:37 am
Hello

I have a 4 class solution for a mixture model with 7 continuous variables and 1 categorical. I have been asked by a statistician to run the model with equal variances (which I believe is the default) and with either equal covariances or varying covariances. I was able to run a model using a WITH statement but in the reading I have done I am unclear as to whether this makes the covariances equal. Can you please clarify?
Also can you help with how I might write the script for varying covariances?

thank you
 Bengt O. Muthen posted on Thursday, June 14, 2018 - 4:52 pm
WITH statements in the Overall part of the model will be held equal across classes. You see this in the results both in terms of the estimates and in TECH1.

Class-varying covariances are obtained when WITH is mentioned in class-specific statements such as:

%c#1%

..WITH ...
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: