Class MannWhitneyUTest

java.lang.Object
org.apache.commons.math3.stat.inference.MannWhitneyUTest

public class MannWhitneyUTest extends Object
An implementation of the Mann-Whitney U test (also called Wilcoxon rank-sum test).
  • Field Details

    • naturalRanking

      private NaturalRanking naturalRanking
      Ranking algorithm.
  • Constructor Details

    • MannWhitneyUTest

      public MannWhitneyUTest()
      Create a test instance using where NaN's are left in place and ties get the average of applicable ranks. Use this unless you are very sure of what you are doing.
    • MannWhitneyUTest

      public MannWhitneyUTest(NaNStrategy nanStrategy, TiesStrategy tiesStrategy)
      Create a test instance using the given strategies for NaN's and ties. Only use this if you are sure of what you are doing.
      Parameters:
      nanStrategy - specifies the strategy that should be used for Double.NaN's
      tiesStrategy - specifies the strategy that should be used for ties
  • Method Details

    • ensureDataConformance

      private void ensureDataConformance(double[] x, double[] y) throws NullArgumentException, NoDataException
      Ensures that the provided arrays fulfills the assumptions.
      Parameters:
      x - first sample
      y - second sample
      Throws:
      NullArgumentException - if x or y are null.
      NoDataException - if x or y are zero-length.
    • concatenateSamples

      private double[] concatenateSamples(double[] x, double[] y)
      Concatenate the samples into one array.
      Parameters:
      x - first sample
      y - second sample
      Returns:
      concatenated array
    • mannWhitneyU

      public double mannWhitneyU(double[] x, double[] y) throws NullArgumentException, NoDataException
      Computes the Mann-Whitney U statistic comparing mean for two independent samples possibly of different length.

      This statistic can be used to perform a Mann-Whitney U test evaluating the null hypothesis that the two independent samples has equal mean.

      Let Xi denote the i'th individual of the first sample and Yj the j'th individual in the second sample. Note that the samples would often have different length.

      Preconditions:

      • All observations in the two samples are independent.
      • The observations are at least ordinal (continuous are also ordinal).

      Parameters:
      x - the first sample
      y - the second sample
      Returns:
      Mann-Whitney U statistic (maximum of Ux and Uy)
      Throws:
      NullArgumentException - if x or y are null.
      NoDataException - if x or y are zero-length.
    • calculateAsymptoticPValue

      private double calculateAsymptoticPValue(double Umin, int n1, int n2) throws ConvergenceException, MaxCountExceededException
      Parameters:
      Umin - smallest Mann-Whitney U value
      n1 - number of subjects in first sample
      n2 - number of subjects in second sample
      Returns:
      two-sided asymptotic p-value
      Throws:
      ConvergenceException - if the p-value can not be computed due to a convergence error
      MaxCountExceededException - if the maximum number of iterations is exceeded
    • mannWhitneyUTest

      public double mannWhitneyUTest(double[] x, double[] y) throws NullArgumentException, NoDataException, ConvergenceException, MaxCountExceededException
      Returns the asymptotic observed significance level, or p-value, associated with a Mann-Whitney U statistic comparing mean for two independent samples.

      Let Xi denote the i'th individual of the first sample and Yj the j'th individual in the second sample. Note that the samples would often have different length.

      Preconditions:

      • All observations in the two samples are independent.
      • The observations are at least ordinal (continuous are also ordinal).

      Ties give rise to biased variance at the moment. See e.g. http://mlsc.lboro.ac.uk/resources/statistics/Mannwhitney.pdf.

      Parameters:
      x - the first sample
      y - the second sample
      Returns:
      asymptotic p-value
      Throws:
      NullArgumentException - if x or y are null.
      NoDataException - if x or y are zero-length.
      ConvergenceException - if the p-value can not be computed due to a convergence error
      MaxCountExceededException - if the maximum number of iterations is exceeded