Version differences PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Anonymous posted on Friday, August 09, 2002 - 6:17 am
Why would a model that ran successfully with
the MLM estimator in version 2.0 NOT run
successfully with either the MUML or MUMLM estimator in version 2.1. Using either the
MUML or MUMLM estimators, one of the latent
variables is undefined; that did not occur
with MLM.
 Linda K. Muthen posted on Friday, August 09, 2002 - 6:50 am
You would need to send the input or output and data to The estimators were not changed. Only their names were changed. I would need to run your problem in both 2.0 and 2.1 to figure out what is happening.
 Anonymous posted on Friday, November 26, 2004 - 11:36 am
I heard from your previous answer that the mplus version 3 (current version )is allowed for multi level SEM with non-recursive model. Now I already have mplus version 2, and want to run that model. Can I do that with version 2?
Thank you very much
 Linda K. Muthen posted on Saturday, November 27, 2004 - 10:23 am
Non-recursive models can be estimated with Version 2.
 Kätlin Peets posted on Wednesday, September 03, 2008 - 6:49 am

A quick question. Why is it so that I get different STDYX estimates when using the Mplus version 3 vs. version 5? They are close, but not identical.
Thank you!
 Linda K. Muthen posted on Wednesday, September 03, 2008 - 7:25 am
It could be that we now standardize residual covariances using residual variances rather than variances. Other than that, I don't know. I would trust the most recent results.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message