|
|
MIMIC model and non-unique solutions |
|
Message/Author |
|
Anonymous posted on Friday, March 02, 2012 - 5:45 am
|
|
|
I am attempting to fit a MIMIC model using the WLSMV estimator and obtain the following error message: THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING PARAMETER 9. I have attempted to supply starting values for the parameter concerned and can obtain convergence. However I have noticed that supplying different starting values gives different solutions. I have tried exploring random starts as I would with mixture modelling but get the following message: *** WARNING in ANALYSIS command The STARTS option is not available with TYPE=GENERAL and ESTIMATOR=WLSMV. STARTS settings will be ignored. Are there workarounds for this issue when fitting a MIMIC model ? Many thanks for any guidance you can offer. |
|
|
Please send the relevant files and your license number to support@statmodel.com. |
|
Back to top |
|
|