Error message involving parameter # PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Marjorie Seaton posted on Tuesday, September 27, 2011 - 12:09 am
I am running 3 reciprocal effects models and have 4 waves of data. Each model is based on 2 of 4 groups. My first model (syntax in next post; 643 cases) includes groups 1 and 3; the second (460 cases), groups 2 and 3; and the third (1101 cases), groups 3 and 4. My MODEL syntax is is similar for all 3 models.

With my first 2 models (but not the third), I get the following error message:
THE MODEL ESTIMATION TERMINATED NORMALLY
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES MAY NOT BE TRUSTWORTHY FOR SOME PARAMETERS DUE TO A NON-POSITIVE DEFINITE FIRST-ORDER DERIVATIVE PRODUCT MATRIX. THIS MAY BE DUE TO THE STARTING VALUES BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE CONDITION NUMBER IS 0.100D-10. PROBLEM INVOLVING PARAMETER 265.
The parameter is not always the same. Can you please advise me where I’m going wrong?
 Marjorie Seaton posted on Tuesday, September 27, 2011 - 12:12 am
This is the syntax referred to in the previous post. (Structural component not included & only 2 time waves shown)


RvHac on MasVog MasVAng SES;
MSCT1 on RvHac MasVog MasVAng SES;
MSCT1 with MACHT1;
MACHT1 on RvHac MasVog MasVAng SES;

MSCT2 on MSCT1 RvHac MACHT1 MasVog MasVAng SES;
MSCT2 with MACHT2;
MACHT2 on RvHac MACHT1 MSCT1 MasVog MasVAng SES;

Where:
RvHac = the group contrast variable (the USEOBSERVATONS command also used)
MasVog and MasVAng = 2 culture variables operationalised using 2 contrasts
SES = continuous variable
MSCT1, MSCT2 = math self-concept times 1 and 2
MACHT1, MACHT2 = math achievement times 1 and 2.
 Linda K. Muthen posted on Tuesday, September 27, 2011 - 9:26 am
Please send the relevant outputs and your license number to support@statmodel.com. Please keep your posts to one window.
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: