Mailing-List: contact user-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Commons Users List" <user@commons.apache.org>
Received-SPF: pass (athena.apache.org: local policy)
Date: Wed, 8 Aug 2012 14:23:57 +0200
From: Gilles Sadowski <gilles@harfang.homelinux.org>
To: user@commons.apache.org
Subject: Re: [Math] Unable to fit a polynomial to data points correctly.
Message-ID: <20120808122357.GL20488@dusk.harfang.homelinux.org>
Mail-Followup-To: user@commons.apache.org
References: 
 <CAD0R3sC2GNVgWJ5oneoR+xC7iaZfNea=xghXUOd-hPi3rLi-fQ@mail.gmail.com>
 <CAD0R3sDssbKaMXqMD+h_RV4j+q9B-SL8KssaiK45hftxb779YQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAD0R3sDssbKaMXqMD+h_RV4j+q9B-SL8KssaiK45hftxb779YQ@mail.gmail.com>
User-Agent: Mutt/1.5.21 (2010-09-15)
X-DCC-scarlet.be-Metrics: sif 20001; Body=1 Fuz1=1 Fuz2=1
X-Virus-Scanned: clamav-milter 0.97.1-exp at sif
X-Virus-Status: Clean
X-Virus-Checked: Checked by ClamAV on apache.org

On Tue, Aug 07, 2012 at 11:59:09PM +0100, Mat Jaggard wrote:
> I have tried to create a polynomial that fits some data using the
> following code:
> 
>                 PolynomialFitter fitter = new PolynomialFitter(14, new
> GaussNewtonOptimizer());
>                 for (int i = 0; i < numValues; i++)
>                 {
>                         fitter.addObservedPoint(xValues[i], yValues[i]);
>                 }
>                 return new PolynomialFunction(fitter.fit());
> 
> I've also tried using a degree of 4. In both cases and using both a
> GaussNewtonOptimizer and a LevenbergMarquardtOptimizer, I'm able to
> get a straight line to be fitted correctly, but the following data
> results with a constant value but very small multipliers for x, and
> higher orders.
> 
> Is anyone able to let me know why this is happening and what I can do about
> it?
> 
> Many thanks,
> Mat.
> 
> The result I get is...
> y = 110.281064 + 0.002316943x - 3.86E-09x^2 + 4.01E-15x^3 - 1.58E-21x^4
> 
> From this data:
> 
> [...]

Plotting those data, and the above polynomial shows that Commons Math did a
decent job. What led you to think that the fit is not correct?
[Maybe the data could be better fitted with another function...]


Regards,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@commons.apache.org
For additional commands, e-mail: user-help@commons.apache.org