MATLAB Central Newsreader - non linear fitting tool

non linear fitting tool

Samoline1 Linke — Fri, 06 Nov 2009 08:32:03 -0500

Hi,

I have read the non linear fitting tool tutorial but could not understand it well. Is there someone who could just tell me the method with the following simple example

e.g.

x = [1 2 3 4 5]

y = [ 0.01 0.02 0.04 0.08 0.16 ]

If we do: scatter (x, y) we can see it would give a look of exponential function.

How can we fit an estimate curve in this graph? I think we need nlintool but how to deal with this data there?

Re: non linear fitting tool

Samoline1 Linke — Fri, 06 Nov 2009 09:54:01 -0500

"Samoline1 Linke" <maganatewoman@yahoo.com> wrote in message <hd0mu3$i27$1@fred.mathworks.com>...
> Hi,
>
> I have read the non linear fitting tool tutorial but could not understand it well. Is there someone who could just tell me the method with the following simple example
>
> e.g.
>
> x = [1 2 3 4 5]
>
> y = [ 0.01 0.02 0.04 0.08 0.16 ]
>
> If we do: scatter (x, y) we can see it would give a look of exponential function.
>
> How can we fit an estimate curve in this graph? I think we need nlintool but how to deal with this data there?
-------------

ADDITION:

I think the formula would be like:

Y = b0 * exp ( b1 * x )

But how to apply it in nlintool?

Re: non linear fitting tool

Tom Lane — Fri, 06 Nov 2009 21:54:46 -0500

> Y = b0 * exp ( b1 * x )
>
> But how to apply it in nlintool?

nlintool(x, y, @(p,x) p(1)*exp(p(2)*x), [.01 .2])

Re: non linear fitting tool

Frederic Moisy — Mon, 09 Nov 2009 10:03:03 -0500

You can use the Ezyfit toolbox for this purpose,
http://www.mathworks.com/matlabcentral/fileexchange/10176-ezyfit-2-30
Have also a look to the example file, where
the fit of an exponential is illustrated:
http://www.mathworks.com/matlabcentral/fx_files/10176/3/content/ezyfit/demo/html/efdemo.html
Frederic

"Samoline1 Linke" <maganatewoman@yahoo.com> wrote in message <hd0mu3$i27$1@fred.mathworks.com>...
> Hi,
>
> I have read the non linear fitting tool tutorial but could not understand it well. Is there someone who could just tell me the method with the following simple example
>
> e.g.
>
> x = [1 2 3 4 5]
>
> y = [ 0.01 0.02 0.04 0.08 0.16 ]
>
> If we do: scatter (x, y) we can see it would give a look of exponential function.
>
> How can we fit an estimate curve in this graph? I think we need nlintool but how to deal with this data there?