Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.graphics.apps.gnuplot > #326

Re: Wrong fit direction / zrange / unreadable graph file

From Ingo Thies <ingo.thies@gmx.de>
Newsgroups comp.graphics.apps.gnuplot
Subject Re: Wrong fit direction / zrange / unreadable graph file
Date 2011-05-19 10:58 +0200
Message-ID <93k4d4Ft5iU1@mid.individual.net> (permalink)
References <01ac512a-8031-41f0-a899-0684927f6c32@n10g2000yqf.googlegroups.com> <iqqpas$cec$1@gwdu112.gwdg.de> <93c9dhFmvmU1@mid.individual.net> <93j3kjFts6U1@mid.dfncis.de>

Show all headers | View raw


On 19.05.2011 01:39, Hans-Bernhard Bröker wrote:

> gnuplot's fit model isn't linear, though.  I.e. "fit" doesn't assume,
> nor does it know, whether the model function is a straight line.  As far
> as gnuplot is concerned, the model is an arbitrarily complicated
> function of one variable and two parameters.

Gnuplot doesn't have to assume that a linear function is to be fitted
since the user already provides its form:

f(x)=a+b*x
fit [rangeinfo] f(x) via a,b

So, no reason for f(x) to behave in a strange way for a=0, or to assume
anything else then exactly the type of function the user is requesting
for. And, as mentioned, there are standard Levenberg-Marquardt
algorithms which do not stumble over that. Therefore, I would guess,
that the problem is caused by the implementation into gnuplot rather
than by the method itself.

-- 
Gruß,
      Ingo

Back to comp.graphics.apps.gnuplot | Previous | NextPrevious in thread | Find similar


Thread

Wrong fit direction / zrange / unreadable graph file Katharina Miwege <k.miwege@googlemail.com> - 2011-05-15 23:59 -0700
  Re: Wrong fit direction / zrange / unreadable graph file Christoph Bersch <usenet@bersch.net> - 2011-05-16 11:04 +0200
    Re: Wrong fit direction / zrange / unreadable graph file Ingo Thies <ingo.thies@gmx.de> - 2011-05-16 11:34 +0200
      Re: Wrong fit direction / zrange / unreadable graph file Christoph Bersch <usenet@bersch.net> - 2011-05-16 11:52 +0200
        Re: Wrong fit direction / zrange / unreadable graph file Richard Owlett <rowlett@pcnetinc.com> - 2011-05-16 06:25 -0500
          Re: Wrong fit direction / zrange / unreadable graph file Christoph Bersch <usenet@bersch.net> - 2011-05-16 14:44 +0200
        Re: Wrong fit direction / zrange / unreadable graph file Ingo Thies <ingo.thies@gmx.de> - 2011-05-16 16:31 +0200
          Re: Wrong fit direction / zrange / unreadable graph file Katharina Miwege <k.miwege@googlemail.com> - 2011-05-16 09:01 -0700
            Re: Wrong fit direction / zrange / unreadable graph file Christoph Bersch <usenet@bersch.net> - 2011-05-17 08:46 +0200
              Re: Wrong fit direction / zrange / unreadable graph file Hans-Bernhard Bröker <HBBroeker@t-online.de> - 2011-05-19 01:34 +0200
          Re: Wrong fit direction / zrange / unreadable graph file Hans-Bernhard Bröker <HBBroeker@t-online.de> - 2011-05-19 01:49 +0200
            Re: Wrong fit direction / zrange / unreadable graph file Ingo Thies <ingo.thies@gmx.de> - 2011-05-19 11:07 +0200
              Re: Wrong fit direction / zrange / unreadable graph file Hans-Bernhard Bröker <HBBroeker@t-online.de> - 2011-05-19 14:02 +0200
      Re: Wrong fit direction / zrange / unreadable graph file Hans-Bernhard Bröker <HBBroeker@t-online.de> - 2011-05-19 01:39 +0200
        Re: Wrong fit direction / zrange / unreadable graph file Ingo Thies <ingo.thies@gmx.de> - 2011-05-19 10:58 +0200

csiph-web