st: moptimize gf evaluator

[Klaus Pforr <kpforr@googlemail.com>]

<>

Dear Listers,

I posted this already a few days ago, but obviously put my question to 
complicated to receive any answers. Is it true, that the gf type 
evaluator for moptimize expects a score matrix of dimension L x K, with 
L being the the number of independet elements and K being the number of 
coefficients? I do not understand how I can derive such a matrix with 
panel data, for which this evaluator type is recommended. Is this a 
matrix of dL[j]/dbeta[i], with L[j] being the log likelihood 
contribution of panel group j derived with respect to beta coeffient i?
best regards

Klaus Pforr

--
__________________________________

Klaus Pforr
MZES AB - A
Universität Mannheim
D - 68131 Mannheim
Tel:  +49-621-181 2801 (nachmittags)
fax:  +49-621-181 2803
URL:  http://www.mzes.uni-mannheim.de

Besucheranschrift: A5, Raum A312
__________________________________

*
*   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/