Ill-conditioned Fisher Matrix PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Arne Floh posted on Thursday, May 17, 2007 - 3:15 pm
I ran the following Mixture Model:

INPUT INSTRUCTIONS


ANALYSIS: TYPE = MIXTURE;
STARTS = 100 10;
STITERATIONS = 20;
MODEL: %OVERALL%
OpC by x1-x3;
OpB by x4-x5;
PSO by x7-x9;
PV by y11-y12;
SAT by y14-y16;
SE by OpC OpB PSO;
ALOY by y6-y8;
WoM by y9-y10;
SAT on SE PV;
PV on SE;
WoM on SE SAT ALOY;
ALOY on SAT;
y17 y18 y19 on ALOY;
y20 on y17-y19;

Unfortunately, I got the following error message:

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ILL-CONDITIONED
FISHER INFORMATION MATRIX. CHANGE YOUR MODEL AND/OR STARTING VALUES.

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO A NON-POSITIVE
DEFINITE FISHER INFORMATION MATRIX. THIS MAY BE DUE TO THE STARTING VALUES
BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE CONDITION
NUMBER IS 0.176D-15.

THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE
COMPUTED. THIS IS OFTEN DUE TO THE STARTING VALUES BUT MAY ALSO BE
AN INDICATION OF MODEL NONIDENTIFICATION. CHANGE YOUR MODEL AND/OR
STARTING VALUES. PROBLEM INVOLVING PARAMETER 55.

Help appreciated. Thx in advance.

Arne
 Linda K. Muthen posted on Thursday, May 17, 2007 - 3:46 pm
This message is related to your particular data and model so it is hard to say anything without more information. Please send your input, data, output, and license number to support@statmodel.com.
 ME posted on Friday, August 24, 2012 - 12:48 pm
I ran the following Mixture Model:

INPUT INSTRUCTIONS

VARIABLE:NAMES ARE u1-u28;
USEVARIABLES ARE u1-u28;
MISSING ARE .;
CLASSES = C(1);
ANALYSIS: TYPE = mixture;
STARTS = 500 100;
MODEL: %OVERALL%
f by u1-u28;
%c#1%
[f*1];
We obtain the following error message:

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ILL-CONDITIONED
FISHER INFORMATION MATRIX. CHANGE YOUR MODEL AND/OR STARTING VALUES.

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO A NON-POSITIVE DEFINITE FISHER INFORMATION MATRIX. THIS MAY BE DUE TO THE STARTING VALUES
BUT MAY ALSO BE AN INDICATION OF MODEL NONIDENTIFICATION. THE CONDITION
NUMBER IS 0.102D-15.

THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE
COMPUTED. THIS IS OFTEN DUE TO THE STARTING VALUES BUT MAY ALSO BE
AN INDICATION OF MODEL NONIDENTIFICATION. CHANGE YOUR MODEL AND/OR STARTING VALUES. PROBLEM INVOLVING PARAMETER 84.

could you please let me know how to fix this issue.
 Linda K. Muthen posted on Friday, August 24, 2012 - 1:15 pm
You cannot free the factor mean. This makes the model not identified.
 ME posted on Friday, August 24, 2012 - 2:57 pm
Could you please let me know how I should modify my syntax to not allow the factor mean to be free?

Thank you for your feedback
 Linda K. Muthen posted on Friday, August 24, 2012 - 3:18 pm
You can't. In cross-sectional models, factor means cannot be identified. They are fixed at zero. Factor means can be identified only in multiple group and multiple time point models.
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: