org.moeaframework.util.sequence
Class Saltelli

java.lang.Object
  org.moeaframework.util.sequence.Saltelli

All Implemented Interfaces:

Sequence

public class Saltelli
extends Object
implements Sequence

Extension of Sobol's sequence for generating the samples for Saltelli's version of Sobol' global variance decomposition. This approach allows the first-, second-, and total-order sensitivities to be computed with lower error rates and fewer samples than other approaches, and is the sequences required for using SobolAnalysis.

The number of samples, N, generated by the generate(int, int) method must be a multiple of 2*D+2, where D is the dimension of the generated samples.

References:

1. Saltelli, A., et al. "Global Sensitivity Analysis: The Primer." John Wiley & Sons Ltd, 2008.

Constructor Summary

Saltelli()
Constructs a Saltelli sequence generator for use in Sobol' global variance decomposition.

Method Summary

double[][]

generate(int N, int D) Returns a N x D matrix of real numbers in the range [0, 1].

Methods inherited from class java.lang.Object

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

Constructor Detail

Saltelli

public Saltelli()

Constructs a Saltelli sequence generator for use in Sobol' global variance decomposition.

Method Detail

generate

public double[][] generate(int N,
                           int D)

Description copied from interface: Sequence

Returns a N x D matrix of real numbers in the range [0, 1].

Specified by:

generate in interface Sequence

Parameters:

N - the number of sample points

D - the dimension of each sample point

Returns:

a N x D matrix of real numbers in the range [0, 1]