org.apache.commons.math3.fitting
Class HarmonicFitter

java.lang.Object
  org.apache.commons.math3.fitting.CurveFitter<HarmonicOscillator.Parametric>
      org.apache.commons.math3.fitting.HarmonicFitter

public class HarmonicFitter
extends CurveFitter<HarmonicOscillator.Parametric>

Class that implements a curve fitting specialized for sinusoids. Harmonic fitting is a very simple case of curve fitting. The estimated coefficients are the amplitude a, the pulsation ω and the phase φ: f (t) = a cos (ω t + φ). They are searched by a least square estimator initialized with a rough guess based on integrals.

Since:

2.0

Version:

$Id: HarmonicFitter.java 7721 2013-02-14 14:07:13Z CardosoP $

Nested Class Summary

static class

HarmonicFitter.ParameterGuesser This class guesses harmonic coefficients from a sample.

Constructor Summary

HarmonicFitter(MultivariateVectorOptimizer optimizer)
Simple constructor.

Method Summary

double[]	fit() Fit an harmonic function to the observed points.
double[]	fit(double[] initialGuess) Fit an harmonic function to the observed points.

Methods inherited from class org.apache.commons.math3.fitting.CurveFitter

addObservedPoint, addObservedPoint, addObservedPoint, clearObservations, fit, fit, getObservations

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

HarmonicFitter

public HarmonicFitter(MultivariateVectorOptimizer optimizer)

Simple constructor.

Parameters:

optimizer - Optimizer to use for the fitting.

Method Detail

fit

public double[] fit(double[] initialGuess)

Fit an harmonic function to the observed points.

Parameters:

initialGuess - First guess values in the following order:

• Amplitude
• Angular frequency
• Phase

Returns:

the parameters of the harmonic function that best fits the observed points (in the same order as above).

fit

public double[] fit()

Fit an harmonic function to the observed points. An initial guess will be automatically computed.

Returns:

the parameters of the harmonic function that best fits the observed points (see the other fit method.

Throws:

NumberIsTooSmallException - if the sample is too short for the the first guess to be computed.

ZeroException - if the first guess cannot be computed because the abscissa range is zero.