Set time scales free PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Anonymous posted on Tuesday, April 30, 2002 - 5:42 pm
As I understand, we may set time scale (e.g., linear or polynomial function) the same for the slope factors across classes. In addition, we may also set the time scale free for some time points, but keep them the same across classes. I am wondering what is the most often used phrase to interpret this kind of parameter setting? For example, we may say "set the order of polynomial function the same across classes" if polynomial function is applied. In the situation where the time scale is set free to estimate, what do we usually say? Just simply say "set the time scale the same across classes" or something else? Thank you very much for your help.
 bmuthen posted on Tuesday, April 30, 2002 - 6:37 pm
I would refer to having free time scores as estimating the shape of the growth curve. This may be done holding the time score estimates equal across classes or not, with corresponding text additions.
 Margaret Roolfe posted on Wednesday, June 17, 2009 - 10:23 pm
I was wondering how to set the free time scores to be unequal or estimated for each class.

I know the default is for the estimated times to be equal across classes.
 Linda K. Muthen posted on Thursday, June 18, 2009 - 2:59 pm
Mention the growth model with free time scores in the class-specif part of the MODEL command.
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: