umontreal.iro.lecuyer.randvar

Class NormalGen

java.lang.Object

umontreal.iro.lecuyer.randvar.RandomVariateGen

umontreal.iro.lecuyer.randvar.NormalGen

Direct Known Subclasses:

NormalACRGen, NormalBoxMullerGen, NormalKindermannRamageGen, NormalPolarGen

public class NormalGen
extends RandomVariateGen

This class implements methods for generating random variates from the normal distribution N(μ, σ). It has mean μ and variance σ², where σ > 0. Its density function is

f (x) = 1/(2π)^1/2σe^{(x-μ)2/(2σ2)}

The nextDouble method simply calls inverseF on the distribution.

The following table gives the CPU time needed to generate 10⁸ standard normal random variates using the different implementations available in SSJ. The first time is for a generator object (non-static method), and the second time is for the static method where no object is created. These tests were made on a machine with processor AMD Athlon 4000, running Red Hat Linux, with clock speed at 2403 MHz. The static method nextDouble() for NormalBoxMullerGen and NormalPolarGen uses only one number out of two that are generated; thus they are twice slower than the non-static method.

Generator	time in seconds	time in seconds
	(object)	(static)
NormalGen	7.67	7.72
NormalACRGen	4.71	4.76
NormalBoxMullerGen	16.07	31.45
NormalPolarGen	7.31	13.74
NormalKindermannRamageGen	5.38	5.34

Constructor Summary

Constructors

Constructor and Description
NormalGen(RandomStream s) Creates a standard normal random variate generator with mean 0 and standard deviation 1, using stream s.
NormalGen(RandomStream s, double mu, double sigma) Creates a normal random variate generator with mean mu and standard deviation sigma, using stream s.
NormalGen(RandomStream s, NormalDist dist) Creates a random variate generator for the normal distribution dist and stream s.

Method Summary

Modifier and Type	Method and Description
double	getMu() Returns the parameter μ of this object.
double	getSigma() Returns the parameter σ of this object.
static double	nextDouble(RandomStream s, double mu, double sigma) Generates a variate from the normal distribution with parameters μ = mu and σ = sigma, using stream s.

Methods inherited from class umontreal.iro.lecuyer.randvar.RandomVariateGen

getDistribution, getStream, nextArrayOfDouble, nextDouble, setStream, toString

Methods inherited from class java.lang.Object

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

Constructor Detail

NormalGen

public NormalGen(RandomStream s,
                 double mu,
                 double sigma)

Creates a normal random variate generator with mean mu and standard deviation sigma, using stream s.

NormalGen

public NormalGen(RandomStream s)

Creates a standard normal random variate generator with mean 0 and standard deviation 1, using stream s.

NormalGen

public NormalGen(RandomStream s,
                 NormalDist dist)

Creates a random variate generator for the normal distribution dist and stream s.

Method Detail

nextDouble

public static double nextDouble(RandomStream s,
                                double mu,
                                double sigma)

Generates a variate from the normal distribution with parameters μ = mu and σ = sigma, using stream s.

getMu

public double getMu()

Returns the parameter μ of this object.

getSigma

public double getSigma()

Returns the parameter σ of this object.