com.opengamma.strata.math.impl.interpolation

Class HermiteCoefficientsProvider

java.lang.Object

com.opengamma.strata.math.impl.interpolation.HermiteCoefficientsProvider

public class HermiteCoefficientsProvider
extends Object

Hermite interpolation is determined if one specifies first derivatives for a cubic interpolant and first and second derivatives for a quintic interpolant.

Constructor Summary

Constructors

Constructor and Description
HermiteCoefficientsProvider()

Method Summary

Modifier and Type	Method and Description
double	endpointDerivatives(double ints1, double ints2, double slope1, double slope2)
double[]	intervalsCalculator(double[] xValues)
double[]	slopesCalculator(double[] yValues, double[] intervals)
double[][]	slopeSensitivityCalculator(double[] intervals) Derivative values of slopes_i with respect to yValues_j, s_{ij}.
double[][]	solve(double[] values, double[] intervals, double[] slopes, double[] first)
double[][]	solve(double[] values, double[] intervals, double[] slopes, double[] first, double[] second)
DoubleMatrix[]	solveWithSensitivity(double[] values, double[] intervals, double[] slopes, double[][] slopeSensitivity, DoubleArray[] firstWithSensitivity)
DoubleMatrix[]	solveWithSensitivity(double[] values, double[] intervals, double[] slopes, double[][] slopeSensitivity, DoubleArray[] firstWithSensitivity, DoubleArray[] secondWithSensitivity)

Methods inherited from class java.lang.Object

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

Constructor Detail

HermiteCoefficientsProvider

public HermiteCoefficientsProvider()

Method Detail

solve

public double[][] solve(double[] values,
                        double[] intervals,
                        double[] slopes,
                        double[] first)

Parameters:

values - (yValues_i)

intervals - (xValues_{i+1} - xValues_{i})

slopes - (yValues_{i+1} - yValues_{i})/(xValues_{i+1} - xValues_{i})

first - First derivatives at xValues_i

Returns:

Coefficient matrix whose i-th row vector is { a_n, a_{n-1}, ...} for the i-th interval, where a_n, a_{n-1},... are coefficients of f(x) = a_n (x-x_i)^n + a_{n-1} (x-x_i)^{n-1} + .... with n=3

solveWithSensitivity

public DoubleMatrix[] solveWithSensitivity(double[] values,
                                           double[] intervals,
                                           double[] slopes,
                                           double[][] slopeSensitivity,
                                           DoubleArray[] firstWithSensitivity)

Parameters:

values - Y values of data

intervals - (xValues_{i+1} - xValues_{i})

slopes - (yValues_{i+1} - yValues_{i})/(xValues_{i+1} - xValues_{i})

slopeSensitivity - Derivative values of slope with respect to yValues

firstWithSensitivity - First derivative values at xValues_i and their yValues dependencies

Returns:

Coefficient matrix and its node dependencies

solve

public double[][] solve(double[] values,
                        double[] intervals,
                        double[] slopes,
                        double[] first,
                        double[] second)

Parameters:

values - (yValues_i)

intervals - (xValues_{i+1} - xValues_{i})

slopes - (yValues_{i+1} - yValues_{i})/(xValues_{i+1} - xValues_{i})

first - First derivatives at xValues_i

second - Second derivatives at xValues_i

Returns:

Coefficient matrix whose i-th row vector is { a_n, a_{n-1}, ...} for the i-th interval, where a_n, a_{n-1},... are coefficients of f(x) = a_n (x-x_i)^n + a_{n-1} (x-x_i)^{n-1} + .... with n=5

solveWithSensitivity

public DoubleMatrix[] solveWithSensitivity(double[] values,
                                           double[] intervals,
                                           double[] slopes,
                                           double[][] slopeSensitivity,
                                           DoubleArray[] firstWithSensitivity,
                                           DoubleArray[] secondWithSensitivity)

Parameters:

values - (yValues_i)

intervals - (xValues_{i+1} - xValues_{i})

slopes - (yValues_{i+1} - yValues_{i})/(xValues_{i+1} - xValues_{i})

slopeSensitivity - Derivative values of slope with respect to yValues

firstWithSensitivity - First derivative values at xValues_i and their yValues dependencies

secondWithSensitivity - Second derivative values at xValues_i and their yValues dependencies

Returns:

Coefficient matrix and its node dependencies

intervalsCalculator

public double[] intervalsCalculator(double[] xValues)

Parameters:

xValues - The x values

Returns:

Intervals of xValues, ( xValues_{i+1} - xValues_i )

slopesCalculator

public double[] slopesCalculator(double[] yValues,
                                 double[] intervals)

Parameters:

yValues - Y values of data

intervals - Intervals of x data

Returns:

( yValues_{i+1} - yValues_i )/( xValues_{i+1} - xValues_i )

slopeSensitivityCalculator

public double[][] slopeSensitivityCalculator(double[] intervals)

Derivative values of slopes_i with respect to yValues_j, s_{ij}.

Parameters:

intervals - Intervals of x data

Returns:

The matrix s_{ij}

endpointDerivatives

public double endpointDerivatives(double ints1,
                                  double ints2,
                                  double slope1,
                                  double slope2)

Parameters:

ints1 - The first interval

ints2 - The second interval

slope1 - The first gradient

slope2 - The second gradient

Returns:

Value of derivative at each endpoint