[6437] in s-news-athena
How to use "gradient" in call to "nls"?
daemon@ATHENA.MIT.EDU (Jacques Rioux)
Tue Feb 14 15:11:43 1995
Date: Tue, 14 Feb 1995 14:36:04 -0500
To: s-news@utstat.toronto.edu
From: jrioux@act.ulaval.ca (Jacques Rioux)
Dear Netters,
Could anyone give me a hint as to how to use the "gradient" argument
in a call to "nls"? I have read the online help and looked at "deriv" and
what I have come to conclude is that gradient should be a matrix of
functions of the parameters that would be reevaluated each time by nls'
iterations. Am I out in left field or am I getting there?
Thanks,
****************************************************************************
* Prof. Jacques Rioux e-mail: jrioux@act.ulaval.ca *
* Ecole d'actuariat tel : (418) 656-7299 *
* Pavillon Vachon *
* Universite Laval fax : (418) 656-7790 *
* Quebec Canada *
* G1K-7P4 *
****************************************************************************
