com.opengamma.strata.math.impl.differentiation

Class ScalarFieldFirstOrderDifferentiator

java.lang.Object

com.opengamma.strata.math.impl.differentiation.ScalarFieldFirstOrderDifferentiator

All Implemented Interfaces:

Differentiator<DoubleArray,Double,DoubleArray>

public class ScalarFieldFirstOrderDifferentiator
extends Object
implements Differentiator<DoubleArray,Double,DoubleArray>

Differentiates a scalar field (i.e. there is a scalar value for every point in some vector space) with respect to the vector space using finite difference.

For a function $y = f(\mathbf{x})$ where $\mathbf{x}$ is a n-dimensional vector and $y$ is a scalar, this class produces a gradient function $\mathbf{g}(\mathbf{x})$, i.e. a function that returns the gradient for each point $\mathbf{x}$, where $\mathbf{g}$ is the n-dimensional vector $\frac{dy}{dx_i}$.

Constructor Summary

Constructors

Constructor and Description
ScalarFieldFirstOrderDifferentiator() Creates an instance using the default values of differencing type (central) and eps (10-5).
ScalarFieldFirstOrderDifferentiator(FiniteDifferenceType differenceType, double eps) Creates an instance that approximates the derivative of a scalar function by finite difference.

Method Summary

Modifier and Type	Method and Description
Function<DoubleArray,DoubleArray>	differentiate(Function<DoubleArray,Double> function) Provides a function that performs the differentiation.
Function<DoubleArray,DoubleArray>	differentiate(Function<DoubleArray,Double> function, Function<DoubleArray,Boolean> domain) Provides a function that performs the differentiation.

Methods inherited from class java.lang.Object

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

Constructor Detail

ScalarFieldFirstOrderDifferentiator

public ScalarFieldFirstOrderDifferentiator()

Creates an instance using the default values of differencing type (central) and eps (10^-5).

ScalarFieldFirstOrderDifferentiator

public ScalarFieldFirstOrderDifferentiator(FiniteDifferenceType differenceType,
                                           double eps)

Creates an instance that approximates the derivative of a scalar function by finite difference.

If the size of the domain is very small or very large, consider re-scaling first. If this value is too small, the result will most likely be dominated by noise. Use around 10^-5 times the domain size.

Parameters:

differenceType - the type, forward, backward or central. In most situations, central is best

eps - the step size used to approximate the derivative

Method Detail

differentiate

public Function<DoubleArray,DoubleArray> differentiate(Function<DoubleArray,Double> function)

Description copied from interface: Differentiator

Provides a function that performs the differentiation.

Specified by:

differentiate in interface Differentiator<DoubleArray,Double,DoubleArray>

Parameters:

function - a function for which to get the differential function

Returns:

a function that calculates the differential

differentiate

public Function<DoubleArray,DoubleArray> differentiate(Function<DoubleArray,Double> function,
                                                       Function<DoubleArray,Boolean> domain)

Description copied from interface: Differentiator

Provides a function that performs the differentiation.

Specified by:

differentiate in interface Differentiator<DoubleArray,Double,DoubleArray>

Parameters:

function - a function for which to get the differential function

domain - a function that returns false if the requested value is not in the domain, true otherwise

Returns:

a function that calculates the differential