From owner-help-octave at bevo dot che dot wisc dot edu  Mon May 19 17:33:13 1997
Subject: Re: Curve Fitting
From: Heber Farnsworth <heberf at u dot washington dot edu>
To: "B. Matthew Knapp" <mattknapp at alum dot mit dot edu>
Cc: help-octave at che dot utexas dot edu
Date: Mon, 19 May 1997 15:32:00 -0700 (PDT)

There is such a function that was contributed by a user to the Matlab ftp
site.  It's in ftp://ftp.mathworks.com/pub/contrib/v4/optim and it's
called curvefit.m.  This fit's straight lines and plots them with the
data.  There is also curvefitnl.m which proports to do the same for a
user-specified non-linear function but I haven't tried it.

If you want to abstract from any particular functional form then I wrote a
function which does kernel smoothing which I could give you.  It fit's the
best curve to 2-dimensional data without any need to specify the
functional form. If you are interested let me know.

  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: http://weber.u.washington.edu/~heberf
  fax:   (206) 685-9392             email: heberf at u dot washington dot edu

On Sun, 18 May 1997, B. Matthew Knapp wrote:

> Is there a way to do curve fits with Octave ?
> similar to the MATLAB  curvefit(x,y,[order])  ?
> I'm using version 1.45
> 
> Great Program! 
> thanks!
> M<
> -- 
> ************************************************************************
> Matt Knapp: Wind Tunnel Test Engineer - Pioneer RocketPlane
> www.dotcon.com (me)		www.rocketplane.com (Pioneer)
> email: mattknapp at alum dot mit dot edu
> pager: 415-423-8095
> Current Residence: San Jose Ca
> ************************************************************************
>