Warnings about starts value PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Ruixue Wang posted on Saturday, October 23, 2010 - 8:16 am
Hi,
There is a warning about my starts value.
2 perturbed starting value run(s) did not converge.
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.
it led to
WARNING: THE BEST LOGLIKELIHOOD VALUE WAS NOT REPLICATED IN 1 OUT OF 1
BOOTSTRAP DRAWS. THE P-VALUE MAY NOT BE TRUSTWORTHY DUE TO LOCAL
MAXIMA. INCREASE THE NUMBER OF RANDOM STARTS USING THE LRTSTARTS OPTION.
How to choose right starts value to avoid local maxima?
Thank you.
 Ruixue Wang posted on Saturday, October 23, 2010 - 3:25 pm
And I increased the number of starts(4 LRTSTARTS are maximum).

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.
Nothing but only this warning showed. Is results trustable with this warning? No other warning about BLRT.
 Linda K. Muthen posted on Sunday, October 24, 2010 - 9:05 am
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: