Known-Class LGM Version differences (... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Alexander Nath posted on Thursday, March 26, 2020 - 12:53 am
Hi,
I have a question regarding Bayes Estimation in Known-Class LGMs with 2 classes. I have calculated Known-Class LGMs on the older version of Mplus (8.3). When the new version (8.4) was released I did the analysis again and encountered a problem. The model fit (PPP and PPC CI) differed significantly between the two versions. Furthermore the estimates of the parameters in the second decimal place were partly different.
So I wanted to ask what is the reason for this?
I hope you can help me!
Many thanks and many greetings
Alexander Nath
 Bengt O. Muthen posted on Thursday, March 26, 2020 - 11:36 am
Version 8.4 PPP has changed as described in our document (see Version History):

http://www.statmodel.com/download/BayesFit.pdf

Differences in the second decimal may be due to not running the Bayes iterations long enough - in either the 8.3 or 8.4 version.
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: