Package org.apache.commons.math4.legacy.stat.descriptive.rank

Class RandomPivotingStrategy

java.lang.Object

org.apache.commons.math4.legacy.stat.descriptive.rank.RandomPivotingStrategy

All Implemented Interfaces:

PivotingStrategy

public class RandomPivotingStrategy
extends Object
implements PivotingStrategy

A strategy of selecting random index between begin and end indices.

Since:

3.4

Constructor Summary

Constructors

Constructor	Description
RandomPivotingStrategy(org.apache.commons.rng.simple.RandomSource randomSource, long seed)	Simple constructor.

Method Summary

Modifier and Type	Method	Description
int	pivotIndex(double[] work, int begin, int end)	Find pivot index of the array so that partition and Kth element selection can be made.

Methods inherited from class java.lang.Object

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

Constructor Detail

RandomPivotingStrategy

public RandomPivotingStrategy(org.apache.commons.rng.simple.RandomSource randomSource,
                              long seed)

Simple constructor.

Parameters:

randomSource - RNG to use for selecting pivot.

seed - Seed for initializing the RNG.

Since:

4.0

Method Detail

pivotIndex

public int pivotIndex(double[] work,
                      int begin,
                      int end)

Find pivot index of the array so that partition and K^th element selection can be made. A uniform random pivot selection between begin and end indices.

Specified by:

pivotIndex in interface PivotingStrategy

Parameters:

work - data array

begin - index of the first element of the slice

end - index after the last element of the slice

Returns:

The index corresponding to a random uniformly selected value between first and the last indices of the array slice

Throws:

MathIllegalArgumentException - MathIllegalArgumentException when indices exceeds range