|
|
 |
| Unpublishable results due to error? |
 
|
| Message/Author |
|
|
|
|
The model we would like to publish contains an error message pertaining to one of our control variables. This is LONGSCAN data, using the study center sites (center1-4) as control variables, and one of them has issues with the Gamma C matrix:
ONE OR MORE MULTINOMIAL LOGIT 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 LATENT VARIABLES AND ANY INDEPENDENT
VARIABLES. THE FOLLOWING PARAMETERS WERE FIXED:
Parameter 67, C#1 ON CENTER4
Parameter 71, C#2 ON CENTER4
Parameter 75, C#3 ON CENTER4
Does this error render the model unusable?
Thank you. |
|
|
|
|
Not necessarily. It may be due to no variation for a control variable dummy in a certain class. If you have a high entropy (say > 0.8) you can do a crosstab between most likely class and the control variable dummies.
If this doesn't help, send output to Support along with license number. |
|
|
| Back to top |
|
|
