Package edu.cnu.mdi.splot.fit

Class PowerLawFitter

java.lang.Object

edu.cnu.mdi.splot.fit.ALeastSquaresFitter

edu.cnu.mdi.splot.fit.PowerLawFitter

All Implemented Interfaces:

IFitStringGetter, IFitter

public final class PowerLawFitter
extends ALeastSquaresFitter

Nonlinear least-squares fitter for a power law with a constant offset:

 y(x) = A * x^n + C
 

Parameter ordering

• params[0] = A — amplitude
• params[1] = n — exponent (real-valued, unconstrained)
• params[2] = C — baseline offset

Typical applications

Log-log scaling laws, stellar flux vs distance, turbulence spectra, fractal relationships, and any data that appears linear on a log-log plot.

Important: domain restriction

The model A * x^n is only defined for x > 0. Any data point with x <= 0 will cause the fit to fail. If your data includes non-positive x values, shift or truncate before fitting.

Initial guess strategy

The baseline C is estimated from the smallest-y points. The offset-subtracted data is log-log linearised: fitting a line to log(y − C) vs log(x) gives n as the slope and log(A) as the intercept.

Author:

heddle

Field Summary

Fields

Modifier and Type	Field	Description
static final int	IDX_A	Index of amplitude parameter A.
static final int	IDX_C	Index of baseline-offset parameter C.
static final int	IDX_N	Index of exponent parameter n.
static final String[]	PARAM_NAMES	Display names used in legend strings and fit summaries.

Fields inherited from class edu.cnu.mdi.splot.fit.ALeastSquaresFitter

CAPSIG, checker, covarianceThreshold, initialGuesser, maxEvaluations, maxIterations, MU, OMEGA, optimizer, PHI, SMALLSIG, SUB0, SUB1, SUB2, subArray, SUBN, SUP0, SUP1, SUP2, SUPN

Constructor Summary

Constructors

Constructor	Description
PowerLawFitter()	Creates a fitter with the default Levenberg-Marquardt optimizer.
PowerLawFitter(org.apache.commons.math3.fitting.leastsquares.LeastSquaresOptimizer optimizer)	Creates a fitter with a custom optimizer.

Method Summary

Modifier and Type	Method	Description
Evaluator	asEvaluator(FitResult fit)	Returns an Evaluator that computes A * x^n + C.
protected double[]	defaultInitialGuess(double[] x, double[] y, double[] weights)	Default initial guess used if the caller does not provide an explicit override.
protected org.apache.commons.math3.fitting.leastsquares.ParameterValidator	defaultValidator()	Default validator/bounds policy (optional).
String	functionForm()
protected int	getParameterCount()
IFitStringGetter	getStringGetter()	Get the descriptive string getter for this fitter.
protected org.apache.commons.math3.fitting.leastsquares.MultivariateJacobianFunction	model(double[] x)	Build the model + analytic (or numeric) Jacobian for this fitter, given x values.
String	modelName()
String	parameterName(int index)	Return the display name of a fitted parameter.

Methods inherited from class edu.cnu.mdi.splot.fit.ALeastSquaresFitter

buildFitResult, clampingValidator, fit, fit, fit, getChecker, getCovarianceThreshold, getMaxEvaluations, getMaxIterations, safeCovariances, validateWeights, validateXY, weightsFromSigmaY

Methods inherited from class java.lang.Object

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

Field Details

IDX_A

public static final int IDX_A

Index of amplitude parameter A.

See Also:

• Constant Field Values

IDX_N

public static final int IDX_N

Index of exponent parameter n.

See Also:

• Constant Field Values

IDX_C

public static final int IDX_C

Index of baseline-offset parameter C.

See Also:

• Constant Field Values

PARAM_NAMES

public static final String[] PARAM_NAMES

Display names used in legend strings and fit summaries.

Constructor Details

PowerLawFitter

public PowerLawFitter()

Creates a fitter with the default Levenberg-Marquardt optimizer.

PowerLawFitter

public PowerLawFitter(org.apache.commons.math3.fitting.leastsquares.LeastSquaresOptimizer optimizer)

Creates a fitter with a custom optimizer.

Parameters:

optimizer - the least-squares optimizer to use (non-null)

Method Details

getParameterCount

protected int getParameterCount()

Specified by:

getParameterCount in class ALeastSquaresFitter

Returns:

number of parameters for this fitter.

model

protected org.apache.commons.math3.fitting.leastsquares.MultivariateJacobianFunction model(double[] x)

Description copied from class: ALeastSquaresFitter

Build the model + analytic (or numeric) Jacobian for this fitter, given x values.

Specified by:

model in class ALeastSquaresFitter

Parameters:

x - x data (length n)

Returns:

a MultivariateJacobianFunction compatible with Commons Math least-squares

defaultInitialGuess

protected double[] defaultInitialGuess(double[] x,
 double[] y,
 double[] weights)

Description copied from class: ALeastSquaresFitter

Default initial guess used if the caller does not provide an explicit override. The default implementation delegates to the ALeastSquaresFitter.initialGuesser.

Subclasses may override if they have a better model-specific guess.

Overrides:

defaultInitialGuess in class ALeastSquaresFitter

Parameters:

x - x data

y - y data

weights - optional weights (may be null)

Returns:

initial parameter vector of length ALeastSquaresFitter.getParameterCount()

defaultValidator

protected org.apache.commons.math3.fitting.leastsquares.ParameterValidator defaultValidator()

Description copied from class: ALeastSquaresFitter

Default validator/bounds policy (optional).

Return null for "no validation". For simple bound constraints, consider using ALeastSquaresFitter.clampingValidator(double[], double[]).

Overrides:

defaultValidator in class ALeastSquaresFitter

Returns:

a parameter validator or null

asEvaluator

public Evaluator asEvaluator(FitResult fit)

Returns an Evaluator that computes A * x^n + C.

Specified by:

asEvaluator in class ALeastSquaresFitter

Parameters:

fit - fit result from a previous call to ALeastSquaresFitter.fit(double[], double[])

Returns:

evaluator for the fitted function

Throws:

NullPointerException - if fit is null

IllegalArgumentException - if the result has the wrong parameter count

modelName

public String modelName()

Returns:

model name suitable for a legend or report

functionForm

public String functionForm()

Returns:

mathematical function form suitable for display

parameterName

public String parameterName(int index)

Description copied from interface: IFitStringGetter

Return the display name of a fitted parameter.

Parameters:

index - zero-based parameter index

Returns:

parameter name

getStringGetter

public IFitStringGetter getStringGetter()

Description copied from class: ALeastSquaresFitter

Get the descriptive string getter for this fitter.

Specified by:

getStringGetter in class ALeastSquaresFitter

Returns:

the string getter