Class AbstractDistribution

java.lang.Object
cern.colt.PersistentObject
cern.jet.random.AbstractDistribution
All Implemented Interfaces:
DoubleFunction, IntFunction, Serializable, Cloneable
Direct Known Subclasses:
AbstractContinousDistribution, AbstractDiscreteDistribution

public abstract class AbstractDistribution extends PersistentObject implements DoubleFunction, IntFunction
Abstract base class for all random distributions. A subclass of this class need to override method nextDouble() and, in rare cases, also nextInt().

Currently all subclasses use a uniform pseudo-random number generation engine and transform its results to the target distribution. Thus, they expect such a uniform engine upon instance construction.

MersenneTwister is recommended as uniform pseudo-random number generation engine, since it is very strong and at the same time quick. makeDefaultGenerator() will conveniently construct and return such a magic thing. You can also, for example, use DRand, a quicker (but much weaker) uniform random number generation engine. Of course, you can also use other strong uniform random number generation engines.

Ressources on the Web:

Check the Web version of the CERN Data Analysis Briefbook . This will clarify the definitions of most distributions.
Also consult the StatSoft Electronic Textbook - the definite web book.

Other useful ressources:

Another site and yet another site describing the definitions of several distributions.
You may want to check out a Glossary of Statistical Terms.
The GNU Scientific Library contains an extensive (but hardly readable) list of definition of distributions.
Use this Web interface to plot all sort of distributions.
Even more ressources: Internet glossary of Statistical Terms, a text book, another text book.
Finally, a good link list Statistics on the Web.

Version:
1.0, 09/24/99
See Also:
  • Field Details

  • Constructor Details

    • AbstractDistribution

      protected AbstractDistribution()
      Makes this class non instantiable, but still let's others inherit from it.
  • Method Details

    • apply

      public double apply(double dummy)
      Equivalent to nextDouble(). This has the effect that distributions can now be used as function objects, returning a random number upon function evaluation.
      Specified by:
      apply in interface DoubleFunction
      Parameters:
      dummy - argument passed to the function.
      Returns:
      the result of the function.
    • apply

      public int apply(int dummy)
      Equivalent to nextInt(). This has the effect that distributions can now be used as function objects, returning a random number upon function evaluation.
      Specified by:
      apply in interface IntFunction
      Parameters:
      dummy - argument passed to the function.
      Returns:
      the result of the function.
    • clone

      public Object clone()
      Returns a deep copy of the receiver; the copy will produce identical sequences. After this call has returned, the copy and the receiver have equal but separate state.
      Overrides:
      clone in class PersistentObject
      Returns:
      a copy of the receiver.
    • getRandomGenerator

      protected RandomEngine getRandomGenerator()
      Returns the used uniform random number generator;
    • makeDefaultGenerator

      public static RandomEngine makeDefaultGenerator()
      Constructs and returns a new uniform random number generation engine seeded with the current time. Currently this is MersenneTwister.
    • nextDouble

      public abstract double nextDouble()
      Returns a random number from the distribution.
    • nextInt

      public int nextInt()
      Returns a random number from the distribution; returns (int) Math.round(nextDouble()). Override this method if necessary.
    • setRandomGenerator

      protected void setRandomGenerator(RandomEngine randomGenerator)
      Sets the uniform random generator internally used.