LCAStandard errors may not be trustwo... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Bridget E Weller posted on Tuesday, April 24, 2007 - 2:17 pm
Im using Mplus to conduct LCA with 3 categorical indicators (2 with 3 options and 1 with 2 options) and 5 continuous indicators. I am able to run a 2 and 4 class solution with no errors; however, when I attempt LCA with 3, 5, etc. classes I get the following error: 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 0.205D-20. PROBLEM INVOLVING PARAMETER 34.

Ive used tech1 to examine the parameter (an have an issues with C#2 parameter) and have changed start values, and yet this problem persist. Would you have any suggests on how I can fix this issues?

Additionally, what are your thoughts about having more classes than variables in the model?

I welcome your comments. Thank you.
 Linda K. Muthen posted on Tuesday, April 24, 2007 - 2:29 pm
With two three-category variables and one binary variable, you have 17 degrees of freedom. You cannot have more than 17 parameters.
 Bridget E Weller posted on Thursday, April 26, 2007 - 11:01 pm
Hi Linda,
Thank you! That makes sense.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message