Class NaturalRanking

java.lang.Object
org.apache.commons.math3.stat.ranking.NaturalRanking
All Implemented Interfaces:
RankingAlgorithm
public class NaturalRanking extends Object implements RankingAlgorithm

Ranking based on the natural ordering on doubles.

NaNs are treated according to the configured NaNStrategy and ties are handled using the selected TiesStrategy. Configuration settings are supplied in optional constructor arguments. Defaults are NaNStrategy.FAILED and TiesStrategy.AVERAGE, respectively. When using TiesStrategy.RANDOM, a RandomGenerator may be supplied as a constructor argument.

Examples:

Input data: (20, 17, 30, 42.3, 17, 50, Double.NaN, Double.NEGATIVE_INFINITY, 17)
NaNStrategyTiesStrategy rank(data)
default (NaNs maximal) default (ties averaged) (5, 3, 6, 7, 3, 8, 9, 1, 3)
default (NaNs maximal) MINIMUM (5, 2, 6, 7, 2, 8, 9, 1, 2)
MINIMAL default (ties averaged) (6, 4, 7, 8, 4, 9, 1.5, 1.5, 4)
REMOVED SEQUENTIAL (5, 2, 6, 7, 3, 8, 1, 4)
MINIMAL MAXIMUM (6, 5, 7, 8, 5, 9, 2, 2, 5)

Since:
2.0

  • Field Details

    • DEFAULT_NAN_STRATEGY

      public static final NaNStrategy DEFAULT_NAN_STRATEGY
      default NaN strategy

    • DEFAULT_TIES_STRATEGY

      public static final TiesStrategy DEFAULT_TIES_STRATEGY
      default ties strategy

    • nanStrategy

      private final NaNStrategy nanStrategy
      NaN strategy - defaults to NaNs maximal

    • tiesStrategy

      private final TiesStrategy tiesStrategy
      Ties strategy - defaults to ties averaged

    • randomData

      private final RandomDataGenerator randomData
      Source of random data - used only when ties strategy is RANDOM

  • Constructor Details

    • NaturalRanking

      public NaturalRanking()
      Create a NaturalRanking with default strategies for handling ties and NaNs.

    • NaturalRanking

      public NaturalRanking(TiesStrategy tiesStrategy)
      Create a NaturalRanking with the given TiesStrategy.
      Parameters:
      tiesStrategy - the TiesStrategy to use

    • NaturalRanking

      public NaturalRanking(NaNStrategy nanStrategy)
      Create a NaturalRanking with the given NaNStrategy.
      Parameters:
      nanStrategy - the NaNStrategy to use

    • NaturalRanking

      public NaturalRanking(NaNStrategy nanStrategy, TiesStrategy tiesStrategy)
      Create a NaturalRanking with the given NaNStrategy and TiesStrategy.
      Parameters:
      nanStrategy - NaNStrategy to use
      tiesStrategy - TiesStrategy to use

    • NaturalRanking

      public NaturalRanking(RandomGenerator randomGenerator)
      Create a NaturalRanking with TiesStrategy.RANDOM and the given RandomGenerator as the source of random data.
      Parameters:
      randomGenerator - source of random data

    • NaturalRanking

      public NaturalRanking(NaNStrategy nanStrategy, RandomGenerator randomGenerator)
      Create a NaturalRanking with the given NaNStrategy, TiesStrategy.RANDOM and the given source of random data.
      Parameters:
      nanStrategy - NaNStrategy to use
      randomGenerator - source of random data

  • Method Details

    • getNanStrategy

      public NaNStrategy getNanStrategy()
      Return the NaNStrategy
      Returns:
      returns the NaNStrategy

    • getTiesStrategy

      public TiesStrategy getTiesStrategy()
      Return the TiesStrategy
      Returns:
      the TiesStrategy

    • rank

      public double[] rank(double[] data)
      Rank data using the natural ordering on Doubles, with NaN values handled according to nanStrategy and ties resolved using tiesStrategy.
      Specified by:
      rank in interface RankingAlgorithm
      Parameters:
      data - array to be ranked
      Returns:
      array of ranks
      Throws:
      NotANumberException - if the selected NaNStrategy is FAILED and a Double.NaN is encountered in the input data

    • removeNaNs

      private NaturalRanking.IntDoublePair[] removeNaNs(NaturalRanking.IntDoublePair[] ranks)
      Returns an array that is a copy of the input array with IntDoublePairs having NaN values removed.
      Parameters:
      ranks - input array
      Returns:
      array with NaN-valued entries removed

    • recodeNaNs

      private void recodeNaNs(NaturalRanking.IntDoublePair[] ranks, double value)
      Recodes NaN values to the given value.
      Parameters:
      ranks - array to recode
      value - the value to replace NaNs with

    • containsNaNs

      private boolean containsNaNs(NaturalRanking.IntDoublePair[] ranks)
      Checks for presence of NaNs in ranks.
      Parameters:
      ranks - array to be searched for NaNs
      Returns:
      true iff ranks contains one or more NaNs

    • resolveTie

      private void resolveTie(double[] ranks, List<Integer> tiesTrace)
      Resolve a sequence of ties, using the configured TiesStrategy. The input ranks array is expected to take the same value for all indices in tiesTrace. The common value is recoded according to the tiesStrategy. For example, if ranks = invalid input: '<'5,8,2,6,2,7,1,2>, tiesTrace = invalid input: '<'2,4,7> and tiesStrategy is MINIMUM, ranks will be unchanged. The same array and trace with tiesStrategy AVERAGE will come out invalid input: '<'5,8,3,6,3,7,1,3>.
      Parameters:
      ranks - array of ranks
      tiesTrace - list of indices where ranks is constant -- that is, for any i and j in TiesTrace, ranks[i] == ranks[j]

    • fill

      private void fill(double[] data, List<Integer> tiesTrace, double value)
      Setsdata[i] = value for each i in tiesTrace.
      Parameters:
      data - array to modify
      tiesTrace - list of index values to set
      value - value to set

    • restoreNaNs

      private void restoreNaNs(double[] ranks, List<Integer> nanPositions)
      Set ranks[i] = Double.NaN for each i in nanPositions.
      Parameters:
      ranks - array to modify
      nanPositions - list of index values to set to Double.NaN

    • getNanPositions

      private List<Integer> getNanPositions(NaturalRanking.IntDoublePair[] ranks)
      Returns a list of indexes where ranks is NaN.
      Parameters:
      ranks - array to search for NaNs
      Returns:
      list of indexes i such that ranks[i] = NaN