Problems with parameter estimation PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 Peter Elliott posted on Thursday, November 28, 2002 - 1:28 pm
When I run a growth mixture model I get figures of 999.000 in the Std and StdYX columns. Everything else looks OK. And if I'd run the program without the command Output: Standardized; I'd never have come across these figures. Does this indicate a problem or am I worrying without cause??

Peter Elliott
 bmuthen posted on Friday, November 29, 2002 - 6:51 am
999 means that it was not possible to compute the quantity. For example, if a variance is zero or negative, dividing by the square root of the variance in a standardization is not possble.
 jbond posted on Tuesday, September 30, 2003 - 1:30 pm
Hello. I am following example 25.1 in the mplus users guide, except it is a 4 class model and there are two continuous outcome variables, each measured at 4 time points. The problem seems to be with identification (the iterations terminate due to an ill-conditioned fisher information matrix). I've tried several combinations even with using the exact analog of example 25.1, with only volume of consumption as the time varying variable (that is, removing the aavariable) but all seemed to have the same problem. The distributions of volume and # aa meetings (applying the USEOBSERVATIONS selection criteria)
although skewed, still have quite a bit of variation. Is there anything obviously wrong in the specification or is it likely a numerical identification problem? Thanks much for any input,

Jason

-----------------------------------

The syntax used was (sorry if this is not the best way to provide the info):

TITLE: LCA For Number of AA Meetings;

DATA:
FILE IS "G:\Trajectories\AA Careers\aacareer.dat";

VARIABLE:
NAMES = id dataset2 White Black Hisp
Gender Age Income dsm4alc dsm4alc6
volcapt1 volcapt2 volcapt3 volcapt4
aapstyr1 aapstyr2 aapstyr3 aapstyr4
havspon1 havspon2 havspon3 havspon4
isspons1 isspons2 isspons3 isspons4
readlit1 readlit2 readlit3 reatlit4
spirawk1 spirawk2 spirawk3 spirawk4
aacapdt1 aacapdt2 aacapdt3 aacapdt4
aasca4t1 aasca4t2 aasca4t3 aasca4t4
aasca2t1 aasca2t2 aasca2t3 aasca2t4
absalct2 absalct3 absalct4
yrfq35t1 yrfq35t2 yrfq35t3 yrfq35t4
rg67 rh67 rq80b rg63 rh63;


USEVARIABLES ARE aacapdt1 aacapdt2 aacapdt3 aacapdt4
volcapt1 volcapt2 volcapt3 volcapt4 id;
USEOBSERVATIONS = dsm4alc EQ 1 and
not(aacapdt1 == 0 and aacapdt2 == 0 and aacapdt3 == 0 and
aacapdt4 == 0);

Classes = C(4);
MISSING ARE ALL (-9);
IDvariable = id;

ANALYSIS:
TYPE = Mixture Missing;

OUTPUT:
TECH1;

MODEL:
%OVERALL%
aacapdt1 with aacapdt2 aacapdt3 aacapdt4;
aacapdt2 with aacapdt3 aacapdt4;
aacapdt3 with aacapdt4;

volcapt1 with volcapt2 volcapt3 volcapt4;
volcapt2 with volcapt3 volcapt4;
volcapt3 with volcapt4;


%C#2%
[aacapdt1-aacapdt4*50];
[volcapt1*2200 volcapt2*930 volcapt3*760 volcapt4*610];

%C#3%
[aacapdt1-aacapdt4*100];
[volcapt1*2200 volcapt2*940 volcapt3*640 volcapt4*590];

%C#4%
[aacapdt1-aacapdt4*200];
[volcapt1*2500 volcapt2*940 volcapt3*780 volcapt4*660];


and the corresponding output is:

PROPORTION OF DATA PRESENT FOR Y


Covariance Coverage
AACAPDT1 AACAPDT2 AACAPDT3 AACAPDT4 VOLCAPT1
________ ________ ________ ________ ________
AACAPDT1 0.980
AACAPDT2 0.709 0.719
AACAPDT3 0.693 0.634 0.703
AACAPDT4 0.634 0.578 0.614 0.644
VOLCAPT1 0.980 0.719 0.703 0.644 1.000
VOLCAPT2 0.733 0.719 0.653 0.594 0.743
VOLCAPT3 0.695 0.632 0.697 0.616 0.705
VOLCAPT4 0.632 0.576 0.612 0.642 0.642


Covariance Coverage
VOLCAPT2 VOLCAPT3 VOLCAPT4
________ ________ ________
VOLCAPT2 0.743
VOLCAPT3 0.653 0.705
VOLCAPT4 0.592 0.614 0.642


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 0.513D-13.

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 21.


FINAL CLASS COUNTS AND PROPORTIONS OF TOTAL SAMPLE SIZE
BASED ON ESTIMATED POSTERIOR PROBABILITIES

Class 1 242.57932 0.49006
Class 2 188.33055 0.38047
Class 3 37.29576 0.07534
Class 4 26.79438 0.05413


CLASSIFICATION OF INDIVIDUALS BASED ON THEIR MOST LIKELY CLASS MEMBERSHIP

Class Counts and Proportions

Class 1 245 0.49495
Class 2 192 0.38788
Class 3 32 0.06465
Class 4 26 0.05253


Average Class Probabilities by Class

1 2 3 4

Class 1 0.911 0.073 0.013 0.003
Class 2 0.091 0.882 0.024 0.003
Class 3 0.045 0.035 0.921 0.000
Class 4 0.014 0.001 0.000 0.985


MODEL RESULTS

Estimates

CLASS 1

AACAPDT1 WITH
AACAPDT2 490.326
AACAPDT3 115.399
AACAPDT4 73.934

AACAPDT2 WITH
AACAPDT3 1719.602
AACAPDT4 845.029

AACAPDT3 WITH
AACAPDT4 438.563

VOLCAPT1 WITH
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT2 WITH
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT3 WITH
VOLCAPT4 *********

Means
AACAPDT1 14.872
AACAPDT2 55.274
AACAPDT3 25.149
AACAPDT4 22.902
VOLCAPT1 1128.434
VOLCAPT2 697.542
VOLCAPT3 653.748
VOLCAPT4 588.990

Variances
AACAPDT1 996.789
AACAPDT2 9602.278
AACAPDT3 2501.427
AACAPDT4 2405.750
VOLCAPT1 *********
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

CLASS 2

AACAPDT1 WITH
AACAPDT2 490.326
AACAPDT3 115.399
AACAPDT4 73.934

AACAPDT2 WITH
AACAPDT3 1719.602
AACAPDT4 845.029

AACAPDT3 WITH
AACAPDT4 438.563

VOLCAPT1 WITH
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT2 WITH
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT3 WITH
VOLCAPT4 *********

Means
AACAPDT1 14.907
AACAPDT2 77.355
AACAPDT3 20.009
AACAPDT4 14.987
VOLCAPT1 3686.941
VOLCAPT2 1085.814
VOLCAPT3 1148.876
VOLCAPT4 801.556

Variances
AACAPDT1 996.789
AACAPDT2 9602.278
AACAPDT3 2501.427
AACAPDT4 2405.750
VOLCAPT1 *********
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

CLASS 3

AACAPDT1 WITH
AACAPDT2 490.326
AACAPDT3 115.399
AACAPDT4 73.934

AACAPDT2 WITH
AACAPDT3 1719.602
AACAPDT4 845.029

AACAPDT3 WITH
AACAPDT4 438.563

VOLCAPT1 WITH
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT2 WITH
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT3 WITH
VOLCAPT4 *********

Means
AACAPDT1 40.630
AACAPDT2 196.997
AACAPDT3 253.225
AACAPDT4 225.177
VOLCAPT1 2686.859
VOLCAPT2 1538.320
VOLCAPT3 563.325
VOLCAPT4 582.173

Variances
AACAPDT1 996.789
AACAPDT2 9602.278
AACAPDT3 2501.427
AACAPDT4 2405.750
VOLCAPT1 *********
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

CLASS 4

AACAPDT1 WITH
AACAPDT2 490.326
AACAPDT3 115.399
AACAPDT4 73.934

AACAPDT2 WITH
AACAPDT3 1719.602
AACAPDT4 845.029

AACAPDT3 WITH
AACAPDT4 438.563

VOLCAPT1 WITH
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT2 WITH
VOLCAPT3 *********
VOLCAPT4 *********

VOLCAPT3 WITH
VOLCAPT4 *********

Means
AACAPDT1 216.803
AACAPDT2 114.169
AACAPDT3 55.023
AACAPDT4 20.042
VOLCAPT1 2505.316
VOLCAPT2 1437.330
VOLCAPT3 1040.784
VOLCAPT4 1229.785

Variances
AACAPDT1 996.789
AACAPDT2 9602.278
AACAPDT3 2501.427
AACAPDT4 2405.750
VOLCAPT1 *********
VOLCAPT2 *********
VOLCAPT3 *********
VOLCAPT4 *********

LATENT CLASS REGRESSION MODEL PART

Means
C#1 2.203
C#2 1.950
C#3 0.331


TECHNICAL 1 OUTPUT


PARAMETER SPECIFICATION FOR CLASS 1


NU
AACAPDT1 AACAPDT2 AACAPDT3 AACAPDT4 VOLCAPT1
________ ________ ________ ________ ________
1 1 2 3 4 5


NU
VOLCAPT2 VOLCAPT3 VOLCAPT4
________ ________ ________
1 6 7 8


THETA
AACAPDT1 AACAPDT2 AACAPDT3 AACAPDT4 VOLCAPT1
________ ________ ________ ________ ________
AACAPDT1 9
AACAPDT2 10 11
AACAPDT3 12 13 14
AACAPDT4 15 16 17 18
VOLCAPT1 0 0 0 0 19
VOLCAPT2 0 0 0 0 20
VOLCAPT3 0 0 0 0 22
VOLCAPT4 0 0 0 0 25


THETA
VOLCAPT2 VOLCAPT3 VOLCAPT4
________ ________ ________
VOLCAPT2 21
VOLCAPT3 23 24
VOLCAPT4 26 27 28


PARAMETER SPECIFICATION FOR CLASS 2


NU
AACAPDT1 AACAPDT2 AACAPDT3 AACAPDT4 VOLCAPT1
________ ________ ________ ________ ________
1 29 30 31 32 33


NU
VOLCAPT2 VOLCAPT3 VOLCAPT4
________ ________ ________
1 34 35 36


THETA
AACAPDT1 AACAPDT2 AACAPDT3 AACAPDT4 VOLCAPT1
________ ________ ________ ________ ________
AACAPDT1 9
AACAPDT2 10 11
AACAPDT3 12 13 14
AACAPDT4 15 16 17 18
VOLCAPT1 0 0 0 0 19
VOLCAPT2 0 0 0 0 20
VOLCAPT3 0 0 0 0 22
VOLCAPT4 0 0 0 0 25


THETA
VOLCAPT2 VOLCAPT3 VOLCAPT4
________ ________ ________
VOLCAPT2 21
VOLCAPT3 23 24
VOLCAPT4 26 27 28
 bmuthen posted on Tuesday, September 30, 2003 - 2:35 pm
You are trying to do a mixture model with class-invariant covariance matrix and class-varying means. In line with Everitt-Hand's book referred to under the Mplus section with classic mixture examples, this is a difficult model to work with (unequal covariance matrices would be even harder). The LPA model in contrast, fixes the off-diagonal covariance matrix elements to zero. I notice that several of your variables are on a very high scale - your analysis might be simpler if you scale your variables down to variances in the 1-10 range.
 Jim Prisciandaro posted on Friday, March 28, 2008 - 1:59 pm
Hello Bengt and Linda,

I have been attempting to estimate a model in line with Example 7.26 (CFA with a non-parametric representation of a non-normal factor distribution) in the Mplus manual. I have tried estimating the model (which has 3 latent classes) exactly as with example 7.26, and I have also attempted to alter the factor means/intercepts for each of the latent classes using contrasts (e.g., -1, 0, 1) or by setting the first mean/intercept to "0" and freely estimating the rest. The models are estimating without error, but all of the models are returning standardized estimates of 999 for all loadings, means, variances, etc. I am aware that 999 = indeterminate, but I am not so sure that 999 = a problem with my models given your post on 11/02 in this thread:

"999 means that it was not possible to compute the quantity. For example, if a variance is zero or negative, dividing by the square root of the variance in a standardization is not possble."

Because Example 7.26 specifies the variance of the latent factor to be 0 in the overall model, does this mean that all models based on this example will not have estimable standardized coefficients? If, alternatively, all of the 999 values are indicative of a problem with my model, are my fit estimates still reliable/interpretable (e.g., BIC) given that standardized estimates are not estimable?

Thanks,
Jim
 Linda K. Muthen posted on Friday, March 28, 2008 - 3:22 pm
The 999's come from the fact that the variance of f is zero. These values are computed after model estimation and do not reflect anything about your model. This does not indicate any other problem.
 Clarissa Wehner posted on Monday, August 24, 2015 - 2:36 am
Hello,

I was trying to run a latent profile analysis for nestled models with six continuous indicator variables. I want to compare models with 2 to 5 classes with different parameter specifications. While using starting values 2000 200, running a model with within-and between-class-varying means and variances, but covariances set to zero worked fine. The only difficulty was that the best loglikelihood-value couldnt be replicated.

When I try running a model with within-varying covariances (but not varying across classes) or a model with varying variances and covariances within and between classes (the most unrestricted model), I get this error message:

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 0.249D-10. PROBLEM INVOLVING PARAMETER 1.
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: {....}
 Clarissa Wehner posted on Monday, August 24, 2015 - 2:37 am
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 0.237D-10.

THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. {...}

In the section "model results" or "model command with final estimates used as starting values", respectively, I received estimates for every parameter. no missing spots or -999-values or ******-symbols. Unfortunately, there is no Tech11, Tech13 and Tech14 Ouput and no model fit statistics like BIC to evaluate my model.

Thank you for your help!
Best regards
Clarissa
 Linda K. Muthen posted on Monday, August 24, 2015 - 5:40 am
Please send your output and license number to support@statmodel.com.

Please limit future 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: