net.sf.doodleproject.numerics4j.statistics.distribution
Class ContinuousDistribution

java.lang.Object
  net.sf.doodleproject.numerics4j.statistics.distribution.ContinuousDistribution

All Implemented Interfaces:

Distribution

Direct Known Subclasses:

BetaDistribution, CauchyDistribution, ChiSquaredDistribution, ExponentialDistribution, FDistribution, GammaDistribution, LaplaceDistribution, LogisticDistribution, LogNormalDistribution, NormalDistribution, RayleighDistribution, TDistribution, UniformDistribution, WeibullDistribution

public abstract class ContinuousDistribution
extends java.lang.Object
implements Distribution

Base continuous distribution.

Version:

$Revision: 1.2 $ $Date: 2007/10/25 04:44:10 $

Constructor Summary

ContinuousDistribution()

Method Summary

abstract double	cumulativeProbability(double x) The CDF for this distribution.
protected double	findInverseCumulativeProbability(double p, double lower, double initial, double upper) Generic means to compute inverse cumulative probability values.
abstract double	inverseCumulativeProbability(double p) The inverse CDF for this distribution.

Methods inherited from class java.lang.Object

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

Constructor Detail

ContinuousDistribution

public ContinuousDistribution()

Method Detail

cumulativeProbability

public abstract double cumulativeProbability(double x)
                                      throws NumericException

The CDF for this distribution. This method returns P(X < x).

Parameters:

x - the value at which the CDF is evaluated.

Returns:

CDF for this distribution.

Throws:

NumericException - if the cumulative probability can not be computed.

inverseCumulativeProbability

public abstract double inverseCumulativeProbability(double p)
                                             throws NumericException

The inverse CDF for this distribution. This method returns x such that, P(X < x) = p.

Parameters:

p - the cumulative probability.

Returns:

x

Throws:

NumericException - if the inverse cumulative probability can not be computed.

findInverseCumulativeProbability

protected double findInverseCumulativeProbability(double p,
                                                  double lower,
                                                  double initial,
                                                  double upper)
                                           throws NumericException

Generic means to compute inverse cumulative probability values. This method uses the bisection method to find inverse CDF values.

Parameters:

p - the cumulative probability.

lower - the global lower bound of the inverse CDF value.

initial - an initial guess at the inverse CDF value.

upper - the global upper bound of the inverse CDF value.

Returns:

x such that P(X < x) = p

Throws:

NumericException - if the inverse cumulative probability can not be computed.