Package jdistlib

Class Multinomial

java.lang.Object
jdistlib.Multinomial
public final class Multinomial extends Object
Multinomial mass and random generation for a vector of category counts.

  • Method Summary

    Modifier and Type
    Method
    Description
    static double
    cumulative(int[] upper, int size, double[] weights)
    Exact lower-orthant probability P(X[i] <= upper[i], all i).
    static double
    density(double[] x, int size, double[] weights, boolean giveLog)
     
    static double
    probability(int[] lower, int[] upper, int size, double[] weights)
    Exact inclusive rectangle probability for the category counts.
    static int[]
    random(int size, double[] weights, RandomEngine random)
     
    static int[]
    random(int size, double[] weights, RandomEngine random, Binomial.RandomState binomialState)
    Draws a multinomial vector using sequential conditional binomials and Kahan compensated probability arithmetic, matching current R's rmultinom.
    static int[][]
    random(int n, int size, double[] weights, RandomEngine random)
     
    static int[][]
    random(int n, int size, double[] weights, RandomEngine random, Binomial.RandomState binomialState)
     

    Methods inherited from class java.lang.Object

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

  • Method Details

    • density

      public static double density(double[] x, int size, double[] weights, boolean giveLog)

    • probability

      public static double probability(int[] lower, int[] upper, int size, double[] weights)
      Exact inclusive rectangle probability for the category counts.

    • cumulative

      public static double cumulative(int[] upper, int size, double[] weights)
      Exact lower-orthant probability P(X[i] <= upper[i], all i).

    • random

      public static int[] random(int size, double[] weights, RandomEngine random)

    • random

      public static int[] random(int size, double[] weights, RandomEngine random, Binomial.RandomState binomialState)
      Draws a multinomial vector using sequential conditional binomials and Kahan compensated probability arithmetic, matching current R's rmultinom. The supplied binomial state also selects corrected or legacy BTPE behavior.

    • random

      public static int[][] random(int n, int size, double[] weights, RandomEngine random)

    • random

      public static int[][] random(int n, int size, double[] weights, RandomEngine random, Binomial.RandomState binomialState)