Non-positive definite Fisher informat... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 FEI Leishi posted on Saturday, January 07, 2017 - 5:21 am
I am doing a two-level path analysis, but the output shows that:

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 -0.833D-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 25.

I have once tried to add the option "STARTS = 0" in the ANALYSIS command, but it does not work.
Do it mean that I'd better change my model?
 FEI Leishi posted on Saturday, January 07, 2017 - 5:26 am
Following my last massage above, part of my input is as following:

[……]
cluster = class;
within = mdadr mdadw mdadp mmomr mmomw mmomp age livemom livedad livegran;
between = justice;
centering = grandmean (mmomr mmomw mmomp mdadr mdadw mdadp);

analysis:
type = twolevel random;

model:
%within%
mtrust on mmomr mmomw mmomp mdadr mdadw mdadp;
mtrust on mesteem;
mtrust on age livemom livedad livegran;
s1 | mesteem on mmomr;
s2 | mesteem on mmomw;
s3 | mesteem on mmomp;
s4 | mesteem on mdadr;
s5 | mesteem on mdadw;
s6 | mesteem on mdadp;

%between%
mesteem on justice;
s1-s6 on justice;

output: TECH1 TECH8 sampstat;
 Linda K. Muthen posted on Saturday, January 07, 2017 - 6:13 am
Please send the input and your license number to support@statmodel.com.
 Yoosun Chu posted on Wednesday, October 04, 2017 - 8:59 am
Hello I encountered the same issue. Would you mind if I send my output and the license number? Thanks.
 Bengt O. Muthen posted on Wednesday, October 04, 2017 - 3:17 pm
Please do.
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: