Latent profile analysis PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 learningmplus posted on Monday, November 11, 2013 - 1:45 pm
Hi,

I am running an LPA with 10 items. The 10 items are peer nominations for specific behavioral indicators (e.g., shy, sad etc.). There are 3913 subjects in the databse and the peer nominations are highly skewed. The program runs, but I get the error below. How can I fix it?

ANDOM STARTS RESULTS RANKED FROM THE BEST TO THE WORST LOGLIKELIHOOD VALUES

Final stage loglikelihood values at local maxima, seeds, and initial stage start numbers:


50 perturbed starting value run(s) did not converge.


THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN INSUFFICIENT
NUMBER OF E STEPS. INCREASE THE NUMBER OF MITERATIONS. ESTIMATES
CANNOT BE TRUSTED.


WARNING: WHEN ESTIMATING A MODEL WITH MORE THAN TWO CLASSES, IT MAY BE
NECESSARY TO INCREASE THE NUMBER OF RANDOM STARTS USING THE STARTS OPTION
TO AVOID LOCAL MAXIMA.
 Linda K. Muthen posted on Monday, November 11, 2013 - 4:40 pm
You can try increasing your random starts to, for example,

STARTS=1000 250;

If you continue to have problems, please send your output and license number to support@statmodel.com.
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: