Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: moptimize gf evaluator

From   Klaus Pforr <>
Subject   st: moptimize gf evaluator
Date   Mon, 15 Nov 2010 15:21:52 +0100


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
Universität Mannheim
D - 68131 Mannheim
Tel:  +49-621-181 2801 (nachmittags)
fax:  +49-621-181 2803

Besucheranschrift: A5, Raum A312

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index