Simulation and Growth Modeling PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Anonymous posted on Thursday, October 06, 2005 - 1:28 pm
I'm trying to simulate a growth model for a grant. I have the model developed the way that I want to determine power. However, I am unable to control the means and variances for the growth factors. For example, I'm trying to constrain these factors of means (2.50 1.75 1.50 .75).

I've tried the following syntax:
[p1@2.50]; p1*10.00;

However, when I put the covariates into the model, these means go away and they are inflated. Without keeping the means in place, I'm unable to be confident in the simulation. How should I proceed with this issue? Any advice you can offer would be great. Thank you very much.
 Linda K. Muthen posted on Saturday, October 08, 2005 - 2:02 pm
When you have covariates in the model, the values in square brackets are no longer means, they are intercepts. You need to take that into account. Likewise, when covariates are added to the model, residual variances and not variances are estimated.
 QianLi Xue posted on Tuesday, August 20, 2013 - 11:12 am
In the simulation file named "mcex6.11.inp" for your EXAMPLE 6.11: PIECEWISE GROWTH MODEL FOR A CONTINUOUS OUTCOME, I did not find the specification of the variance and covariance parameters for random growth factors: i,s1,s2.What are your default values for them? Thanks!
 Linda K. Muthen posted on Tuesday, August 20, 2013 - 1:24 pm
The variances are shown as

i*1; s1-s2*.2;

The covariance is not shown so the value zero is used. Any parameter for which a population value is not shown is given the value zero.
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: