jdistlib

Class NonCentralChiSquare

java.lang.Object

jdistlib.NonCentralChiSquare

public class NonCentralChiSquare
extends Object

Constructor Summary

Constructors

Constructor and Description
NonCentralChiSquare()

Method Summary

Methods

Modifier and Type	Method and Description
static double	cumulative_raw(double x, double f, double theta, double errmax, double reltol, int itrmax, boolean lower_tail)
static double	cumulative(double x, double df, double ncp, boolean lower_tail, boolean log_p)
static double	density(double x, double df, double ncp, boolean give_log)
static double	quantile(double p, double df, double ncp, boolean lower_tail, boolean log_p)
static double	random(double df, double lambda, QRandomEngine random) According to Hans R.

Methods inherited from class java.lang.Object

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

Constructor Detail

NonCentralChiSquare

public NonCentralChiSquare()

Method Detail

density

public static final double density(double x,
             double df,
             double ncp,
             boolean give_log)

cumulative_raw

public static final double cumulative_raw(double x,
                    double f,
                    double theta,
                    double errmax,
                    double reltol,
                    int itrmax,
                    boolean lower_tail)

cumulative

public static final double cumulative(double x,
                double df,
                double ncp,
                boolean lower_tail,
                boolean log_p)

quantile

public static final double quantile(double p,
              double df,
              double ncp,
              boolean lower_tail,
              boolean log_p)

random

public static final double random(double df,
            double lambda,
            QRandomEngine random)

         According to Hans R. Kuensch's suggestion (30 sep 2002):

          It should be easy to do the general case (ncp > 0) by decomposing it
          as the sum of a central chisquare with df degrees of freedom plus a
          noncentral chisquare with zero degrees of freedom (which is a Poisson
          mixture of central chisquares with integer degrees of freedom),
          see Formula (29.5b-c) in Johnson, Kotz, Balakrishnan (1995).

          The noncentral chisquare with arbitary degrees of freedom is of interest
          for simulating the Cox-Ingersoll-Ross model for interest rates in
          finance.

          R code that works is

            rchisq0 <- function(n, ncp) {
                p <- 0 < (K <- rpois(n, lambda = ncp / 2))
                r <- numeric(n)
                r[p] <- rchisq(sum(p), df = 2*K[p])
                r
            }

            rchisq <- function(n, df, ncp=0) {
                if(missing(ncp)) .Internal(rchisq(n, df))
                else rchisq0(n, ncp) + .Internal(rchisq(n, df))
            }