Error Message - non convergence in MSEM PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Rachel Frieder posted on Wednesday, July 27, 2016 - 12:41 pm
Hello. I am getting the following error messages (where SUPPS represents a L2 variable). Is this because I don't have any between level variance? I was wondering how you recommend that I proceed? I am trying to test a 1-(1-1)-1 and a 2-(1-1)-1. I attempted to do this with latent variables and observed variables and got the same error message.

THE ESTIMATED WITHIN COVARIANCE MATRIX IS NOT POSITIVE DEFINITE AS IT SHOULD BE. COMPUTATION COULD NOT BE COMPLETED.
THE VARIANCE OF SUPPS APPROACHES 0. FIX THIS VARIANCE AND THE CORRESPONDING COVARIANCES TO 0, DECREASE THE MINIMUM VARIANCE, OR SPECIFY THE VARIABLE AS A BETWEEN VARIABLE.

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ERROR IN THE COMPUTATION. CHANGE YOUR MODEL AND/OR STARTING VALUES.

A MATRIX COULD NOT BE INVERTED DURING THE H1 MODEL ESTIMATION. THE ESTIMATED WITHIN COVARIANCE MATRIX IS NOT POSITIVE DEFINITE AS IT SHOULD BE. COMPUTATION COULD NOT BE COMPLETED. THE VARIANCE OF SUPPS APPROACHES 0. FIX THIS VARIANCE AND THE CORRESPONDING COVARIANCES TO 0, DECREASE THE MINIMUM VARIANCE, OR SPECIFY THE VARIABLE AS A BETWEEN VARIABLE.

THE H1 MODEL ESTIMATION DID NOT CONVERGE. CHI-SQUARE TEST AND SAMPLE STATISTICS COULD NOT BE COMPUTED.
 Rachel Frieder posted on Wednesday, July 27, 2016 - 1:19 pm
As a follow up to the above, I remembered that I needed to specify SUPPS as a between variable. So after specifying that Between = SUPPS;

The model ran, and I got output in addition to the following error message:

THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO A CHANGE IN THE LOGLIKELIHOOD DURING THE LAST E STEP.

AN INSUFFICENT NUMBER OF E STEP ITERATIONS MAY HAVE BEEN USED. INCREASE THE NUMBER OF MITERATIONS OR INCREASE THE MCONVERGENCE VALUE. ESTIMATES CANNOT BE TRUSTED. SLOW CONVERGENCE DUE TO PARAMETER 12.

THE LOGLIKELIHOOD DERIVATIVE FOR THIS PARAMETER IS 0.46693364D+00.


How should I proceed with this new error message (note that I no longer receive the "not positive definite" error!). Thanks so much!
 Bengt O. Muthen posted on Wednesday, July 27, 2016 - 2:18 pm
Please send your output and license number to Support.
 Sisi Li posted on Sunday, June 17, 2018 - 11:37 am
Hi,
I'm quite new with Mplus, and I got similar error message when I tried to run a dyadic mediation (1-1-1). My X variables are CL and CLpc, my mediator is Psatpc, and my Y variable is CL; all continuous and measured at level 1. Below is my code and the error message, how should I proceed with this error message? Thanks a lot for helping!

MODEL:%WITHIN%
Asa|Psatpc on CLpc;!AX on M
Asc|CL on CLpc;!AX on Y
PAsa|Psatpc on PCLpc;!PX on M
Psc|CL on PCLpc;!PX on Y
sb|CL on Psatpc;!M on Y
CLpc WITH PCLpc;!correlation of AX & PX
%BETWEEN%
[Asa](Aa);!mean of Asa
[PAsa](Pa);!mean of PAsa
[Asc](Ac);!mean of Asc
[Psc](Pc);!mean of Psc
[sb](b);!mean of sb
Asa WITH sb(cAsab);!covariance between random slopes
PAsa WITH sb(cPAsab);
Asa WITH Asc Psc;
sb WITH Asc Psc;
MODEL CONSTRAINT:
NEW(direct indirect1 indirect2);
indirect1=Aa*b+cAsab;
indirect2=Pa*b+cPAsab;
direct=Ac+Pc;
 Bengt O. Muthen posted on Monday, June 18, 2018 - 9:41 am
Please send your output to Support along with your license number.
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: