Ill-conditioned Fisher information ma... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Sara Casalin posted on Wednesday, May 16, 2012 - 5:33 am
Iíve just started working with Mplus, and am thus fairly new at this. Iím running a multilevel SEM, but keep getting 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 0.337D-14.
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 47.

Any help/suggestions would be much appreciated!
 Linda K. Muthen posted on Wednesday, May 16, 2012 - 6:02 am
Please send your output and license number to support@statmodel.com.
 Anne-Lise Demortier posted on Thursday, October 04, 2012 - 9:32 am
Dear Dr. Muthen,
we are trying to test three multilevel mediation models (SEM). The models begin to run, but we finally get the following error message for each of them:

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 -0.377D-16.

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 62.


THE H1 MODEL ESTIMATION DID NOT CONVERGE. SAMPLE STATISTICS COULD NOT
BE COMPUTED. INCREASE THE NUMBER OF H1ITERATIONS.

Could you please help us to solve this problem?
Thank you very much

Anne-Lise
 Linda K. Muthen posted on Thursday, October 04, 2012 - 9:58 am
Please send the output and your license number to support@statmodel.com.
 Lavinia Flückiger posted on Thursday, January 09, 2014 - 5:14 am
Dear authors

I am running a multilevel SEM mediation model but I always get the following error message, too.

ONE OR MORE PARAMETERS WERE FIXED TO AVOID SINGULARITY OF THE
INFORMATION MATRIX. THE SINGULARITY IS MOST LIKELY BECAUSE THE
MODEL IS NOT IDENTIFIED, OR BECAUSE OF EMPTY CELLS IN THE JOINT
DISTRIBUTION OF THE CATEGORICAL VARIABLES IN THE MODEL.
THE FOLLOWING PARAMETERS WERE FIXED:
24 25 26 29 30

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 0.157D-11.

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 17.

Do you have any ideas to solve this problem?

Thank you very much for your help!

Kind regards, Lav
 Linda K. Muthen posted on Thursday, January 09, 2014 - 6:44 am
Please send the output and your license number to support@statmodel.com.
 Holly Andrewes posted on Thursday, December 04, 2014 - 10:08 pm
Dear Dr Muthen,

I am trying to run two quadratic growth models with individually varying times of observation to assess changes before and after an event which occurs in one group (group1) but not the other (group0). I have 5 time points (T1, T2, T3, T4, T5) and have centred time 0 at T3 when the event occurs for group 1.

When I run the first model I get the output:
THE ESTIMATED COVARIANCE MATRIX IS NOT POSITIVE DEFINITE AS IT SHOULD BE. COMPUTATION COULD NOT BE COMPLETED. PROBLEM INVOLVING VARIABLE S. THE RESIDUAL CORRELATION BETWEEN Q AND S IS -1.084. THE PROBLEM MAY BE RESOLVED BY SETTING ALGORITHM=EM AND MCONVERGENCE TO A LARGE VALUE.

When I run the second model I get the output:

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 0.209D-16.

I am hoping you might know how to solve this problem?
I am also hoping to develop panel plots or spaghetti plots to assess individual change overtime but cant seem to find any code for this. Would it be possible to direct me to a resource that provides this code?

Best wishes,
Holly
 Linda K. Muthen posted on Friday, December 05, 2014 - 5:54 am
Please send the two outputs 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: