IRPact_modellierung.distributions

Class PoissonDistribution

java.lang.Object

IRPact_modellierung.distributions.Distribution

IRPact_modellierung.distributions.UnivariateDistribution

IRPact_modellierung.distributions.DiscreteDistribution

IRPact_modellierung.distributions.COLTDiscreteDistribution

IRPact_modellierung.distributions.PoissonDistribution

public class PoissonDistribution
extends COLTDiscreteDistribution

Class to model a poisson distribution using the COLT distribution library with an average number of events k. Will find the number of events that correspond to the cumulative distribution with respect to a uniformly distributed variable

Field Summary

Fields inherited from class IRPact_modellierung.distributions.COLTDiscreteDistribution

generator

Fields inherited from class IRPact_modellierung.distributions.Distribution

name

Constructor Summary

Constructors

Constructor and Description
PoissonDistribution(java.lang.String name, double lambda) Models a Poisson distribution with an average number of lambda events.

Method Summary

Modifier and Type	Method and Description
double	draw() Draw a value from the distribution according to the distribution at hand
double	getLambda()
double	testDistribution(int numberOfDraws) Method to test whether my assumption of what the COLT implementation does.

Methods inherited from class IRPact_modellierung.distributions.Distribution

getName, toString

Methods inherited from class java.lang.Object

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

Constructor Detail

PoissonDistribution

public PoissonDistribution(java.lang.String name,
                           double lambda)
                    throws java.lang.IllegalArgumentException

Models a Poisson distribution with an average number of lambda events.

Parameters:

name - Name of the distribution (will be prefixed with PoissonDistribution_)

lambda - Average number of each draw.

Throws:

java.lang.IllegalArgumentException - will be thrown if the average number of events (lambda) is negative

Method Detail

draw

public double draw()

Description copied from class: UnivariateDistribution

Draw a value from the distribution according to the distribution at hand

Specified by:

draw in class UnivariateDistribution

Returns:

A double number drawn from the modeled distribution

getLambda

public double getLambda()

testDistribution

public double testDistribution(int numberOfDraws)

Method to test whether my assumption of what the COLT implementation does. Should return something close to lambda (converging with number of draws)

Parameters:

numberOfDraws - number of samples taken

Returns:

the average number of events with numberOfDraws draws