distributions
Class CDF_Normal_Amos

java.lang.Object
  |
  +--distributions.CDF_Normal_Amos

public class CDF_Normal_Amos
extends java.lang.Object

This class contains a Java translation of a FORTRAN routine written by Don Amos and S.L. Daniel to calculate the normal cumulative distribution function (CDF).


Constructor Summary
CDF_Normal_Amos()
           
 
Method Summary
static double fnorm(double x, int kode, int[] nz)
          This method calculates the normal cumulative distribution function.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

CDF_Normal_Amos

public CDF_Normal_Amos()
Method Detail

fnorm

public static double fnorm(double x,
                           int kode,
                           int[] nz)
This method calculates the normal cumulative distribution function. It is a Java translation of the FORTRAN routine fnorm written by Don Amos and S.L. Daniel. The documentation in the FORTRAN code states that "significant digits are retained over the full exponent range with a nominal accuracy max(unit roundoff,1.0e-13)." For IEEE 754 double precision arithmetic, this max value is 1.0e-13.

Steve Verrill translated the FORTRAN code (dated October, 1974) into Java. This translation was performed on February 7, 2001.

Parameters:
x - If kode = 1, the method returns the value of the normal cumulative distribution function at x. If kode = 2, the method returns 1 minus the value of the normal cumulative distribution function at x.
kode -