[Top][All Lists]

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

Re: here is a parametric version of fsolve

From: Craig Earls
Subject: Re: here is a parametric version of fsolve
Date: Tue, 04 Mar 1997 21:37:13 -0500

Heber Farnsworth wrote:
> Craig,
>         Thanks for the fsolve replacement.  I haven't looked closely at it
> but I have a question.  Can the parameters be matrices?  The reason I'm
> asking is that my research is in statistics/econometrics and I frequently
> need to use fsolve to fit models to data.  I need to pass starting values
> to fsolve of course but I need to be able to pass one or more matrices of
> data as well.  The data is what you call parameters, that is the data
> determines the function that fsolve works on but the data cannot be
> changed by fsolve.
>   Heber Farnsworth                               | Department of Finance
>   Univerity of Washington                        | Box 353200
>   tele:  (206) 528-0793 home                     | Seattle, WA 98195-3200
>   tele:  (206) 543-4773 finance     web:
>   fax:   (206) 685-9392             email: address@hidden

The extra parameters can be any legal octave value, the new fsolve
simply works by storing all the arguments after the independent variable
then passing them to foo separately. I have only tried it with scalars,
but I can't see a reason matrices won't work. I actually only had to add
about 10 lines to fsolve. Let me know it it has problems.
Craig P Earls                                 address@hidden
LT US Navy, MIT Ocean Engineering             address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]