jhplot.math

Class KolmogorovSmirnovDist

java.lang.Object

jhplot.math.KolmogorovSmirnovDist

public class KolmogorovSmirnovDist
extends Object

This class computes both the cumulative probability P[D_n <= x] and the complementary cumulative probability P[D_n >= x] of the 2-sided 1-sample Kolmogorov-Smirnov distribution.

The Kolmogorov-Smirnov test statistic D_n is defined by

D_n = sup_x |F(x) - S_n(x)|

where n is the sample size, S_n(x) is an empirical distribution function, and F(x) is a completely specified theoretical distribution.

Since:

1 March 2010

Constructor Summary

Constructors

Constructor and Description
KolmogorovSmirnovDist()

Method Summary

Modifier and Type	Method and Description
static double	cdf(int n, double x) Computes the cumulative probability P[D_n <= x] of the Kolmogorov-Smirnov distribution with sample size n at x.
static double	fbar(int n, double x) Computes the complementary cumulative probability P[D_n >= x] of the Kolmogorov-Smirnov distribution with sample size n at x.
static void	main(String[] args)

Methods inherited from class java.lang.Object

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

Constructor Detail

KolmogorovSmirnovDist

public KolmogorovSmirnovDist()

Method Detail

cdf

public static double cdf(int n,
                         double x)

Computes the cumulative probability P[D_n <= x] of the Kolmogorov-Smirnov distribution with sample size n at x. It returns at least 13 decimal digits of precision for n <= 140, at least 5 decimal digits of precision for 140 < n <= 100000, and a few correct decimal digits for n > 100000.

Parameters:

n - sample size

x - value of Kolmogorov-Smirnov statistic

Returns:

cumulative probability

fbar

public static double fbar(int n,
                          double x)

Computes the complementary cumulative probability P[D_n >= x] of the Kolmogorov-Smirnov distribution with sample size n at x. It returns at least 10 decimal digits of precision for n <= 140, at least 5 decimal digits of precision for 140 < n <= 200000, and a few correct decimal digits for n > 200000.

Parameters:

n - sample size

x - value of Kolmogorov-Smirnov statistic

Returns:

complementary cumulative probability

main

public static void main(String[] args)