Package com.irurueta.numerical

Class DerivativeEstimator

java.lang.Object

com.irurueta.numerical.DerivativeEstimator

Direct Known Subclasses:

SavitzkyGolayDerivativeEstimator, SymmetricDerivativeEstimator

public class DerivativeEstimator
extends Object

Class to estimate the derivative of a single dimension function at a given point. The algorithm used in this implementation is valid for continuous functions only, otherwise inaccurate results might be obtained. This implementation is faster although less accurate than SymmetricDerivativeEstimator.

Field Summary

Fields

Modifier and Type	Field	Description
static final double	EPS	Constant defining machine precision for this algorithm.
protected final SingleDimensionFunctionEvaluatorListener	listener	Listener to evaluate a single dimension function.

Constructor Summary

Constructors

Constructor	Description
DerivativeEstimator(SingleDimensionFunctionEvaluatorListener listener)	Constructor

Method Summary

Modifier and Type	Method	Description
double	derivative(double x)	Computes the function derivative at provided point x.

Methods inherited from class java.lang.Object

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

Field Details

EPS

public static final double EPS

Constant defining machine precision for this algorithm.

See Also:

• Constant Field Values

listener

protected final SingleDimensionFunctionEvaluatorListener listener

Listener to evaluate a single dimension function.

Constructor Details

DerivativeEstimator

public DerivativeEstimator(SingleDimensionFunctionEvaluatorListener listener)

Constructor

Parameters:

listener - listener to evaluate a single dimension function

Method Details

derivative

public double derivative(double x)
                  throws EvaluationException

Computes the function derivative at provided point x.

Parameters:

x - Point where derivative is estimated

Returns:

Derivative of function at provided point

Throws:

EvaluationException - Raised if function cannot be properly evaluated