Variance across classes PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Blair Beadnell posted on Monday, April 24, 2006 - 11:22 am
As i understand, in mixture modeling with continuous variables, the variances are by default constrained to be equal across classes (and can be freed by mentioning them in the model). Hence, the classes should be equally homogenous with each other.

Is there any similar type of default assumption when doing mixture modeling with ordered categorical indicators (using the 'CATEGORICAL = command')?
 Linda K. Muthen posted on Monday, April 24, 2006 - 12:15 pm
There are no variances with categorocal outcomes so this is not relevant. Thresholds are not held equal across class 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: