Ada Lo wrote:
<snip>
I'd agree with Maarten, your model looks very complex, probably too complex. That's a lot of random effects.
>>Now, using varZ and varY, in the exact same command as above, I got
the output below. What does Hessian is not negative semidefinite
really mean?
Hessian not negative-semi definite is bad. This means that the information matrix is not a valid information matrix, which is why you're not getting standard errors. Now the SEs on random effects are kind of misleading anyway but the main use has happened: they show when your model has blown up.
>What can I do about my data to make the model converge?
Probably nothing with the data, you'll need to simplify the model or go Bayesian.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
