Higher-order Multivariate growth curv... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Kara Thompson posted on Monday, January 07, 2013 - 5:28 pm
Hi,

I am trying to reparameterize a multivariate piecewise growth model (with 3 alcohol use measures) into a higher-order factor of curves model, but I continue to get an error message ...NOT POSITIVE DEFINITE." because one of my first-order latent variables has a negative residual variance, but I am not sure how to fix this.

Thanks. Please find my syntax in a subsequent post
 Kara Thompson posted on Monday, January 07, 2013 - 5:30 pm
Model
ib s1b | ...;
ib s2b | ...;
if s1f | ...;
if s2f | ...;
iq s1q | ...;
iq s2q | ...;
itot by if* (1);
itot by iq* (2);
s1tot by s1f* (1);
s1tot by s1q* (2);
s2tot by s2f* (1);
s2tot by s2q* (2);
itot by ib@1;
s1tot by s1b@1;
s2tot by s2b@1;

[freq15-y25@0];
[iq s1q s2q if s1f s2f ib@0 s1b@0 s2b@0];

q25@0;
itot with s1tot;
itot with s2tot;
s1tot with s2tot;

[itot s1tot s2tot];
 Linda K. Muthen posted on Monday, January 07, 2013 - 6:09 pm
Please send the output and your license number to support@statmodel.com.
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: