cern.jet.random.engine
Class RandomEngine

java.lang.Object
  |
  +--cern.colt.PersistentObject
        |
        +--edu.cornell.lassp.houle.RngPack.RandomElement
              |
              +--edu.cornell.lassp.houle.RngPack.RandomSeedable
                    |
                    +--cern.jet.random.engine.RandomEngine
All Implemented Interfaces:
java.lang.Cloneable, cern.colt.function.DoubleFunction, cern.colt.function.IntFunction, java.io.Serializable
Direct Known Subclasses:
DRand, MersenneTwister

public abstract class RandomEngine
extends RandomSeedable
implements cern.colt.function.DoubleFunction, cern.colt.function.IntFunction

Abstract base class for uniform pseudo-random number generating engines.

Most probability distributions are obtained by using a uniform pseudo-random number generation engine (derived from this class or RandomElement) followed by a transformation to the desired distribution. Thus, subclasses of this class are at the core of computational statistics, simulations, Monte Carlo methods, etc.

Subclasses produce uniformly distributed int's and long's in the closed intervals [Integer.MIN_VALUE,Integer.MAX_VALUE] and [Long.MIN_VALUE,Long.MAX_VALUE], respectively, as well as float's and double's in the open unit intervals (0.0f,1.0f) and (0.0,1.0), respectively.

Subclasses need to override one single method only: nextInt(). All other methods generating different data types or ranges are usually layered upon nextInt(). long's are formed by concatenating two 32 bit int's. float's are formed by dividing the interval [0.0f,1.0f] into 232 sub intervals, then randomly choosing one subinterval. double's are formed by dividing the interval [0.0,1.0] into 264 sub intervals, then randomly choosing one subinterval.

Note that this implementation is not synchronized.

See Also:
MersenneTwister, MersenneTwister64, Ranmar, Ranlux, Ranecu, Random, Serialized Form

Constructor Summary
protected RandomEngine()
          Makes this class non instantiable, but still let's others inherit from it.
 
Method Summary
 double apply(double dummy)
          Equivalent to raw().
 int apply(int dummy)
          Equivalent to nextInt().
static RandomElement makeDefault()
          Constructs and returns a new uniform random number engine seeded with the current time.
 double nextDouble()
          Returns a 64 bit uniformly distributed random number in the open unit interval (0.0,1.0) (excluding 0.0 and 1.0).
 float nextFloat()
          Returns a 32 bit uniformly distributed random number in the open unit interval (0.0f,1.0f) (excluding 0.0f and 1.0f).
abstract  int nextInt()
          Returns a 32 bit uniformly distributed random number in the closed interval [Integer.MIN_VALUE,Integer.MAX_VALUE] (including Integer.MIN_VALUE and Integer.MAX_VALUE);
 long nextLong()
          Returns a 64 bit uniformly distributed random number in the closed interval [Long.MIN_VALUE,Long.MAX_VALUE] (including Long.MIN_VALUE and Long.MAX_VALUE).
 double raw()
          Returns a 32 bit uniformly distributed random number in the open unit interval (0.0,1.0) (excluding 0.0 and 1.0).
 
Methods inherited from class edu.cornell.lassp.houle.RngPack.RandomSeedable
ClockSeed, ClockSeed
 
Methods inherited from class edu.cornell.lassp.houle.RngPack.RandomElement
choose, choose, gaussian, gaussian, powlaw, raw, raw, uniform
 
Methods inherited from class cern.colt.PersistentObject
clone
 
Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

RandomEngine

protected RandomEngine()
Makes this class non instantiable, but still let's others inherit from it.

Method Detail

apply

public double apply(double dummy)
Equivalent to raw(). This has the effect that random engines can now be used as function objects, returning a random number upon function evaluation.

Specified by:
apply in interface cern.colt.function.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 random engines can now be used as function objects, returning a random number upon function evaluation.

Specified by:
apply in interface cern.colt.function.IntFunction
Parameters:
dummy - argument passed to the function.
Returns:
the result of the function.

makeDefault

public static RandomElement makeDefault()
Constructs and returns a new uniform random number engine seeded with the current time. Currently this is MersenneTwister.


nextDouble

public double nextDouble()
Returns a 64 bit uniformly distributed random number in the open unit interval (0.0,1.0) (excluding 0.0 and 1.0).


nextFloat

public float nextFloat()
Returns a 32 bit uniformly distributed random number in the open unit interval (0.0f,1.0f) (excluding 0.0f and 1.0f).


nextInt

public abstract int nextInt()
Returns a 32 bit uniformly distributed random number in the closed interval [Integer.MIN_VALUE,Integer.MAX_VALUE] (including Integer.MIN_VALUE and Integer.MAX_VALUE);


nextLong

public long nextLong()
Returns a 64 bit uniformly distributed random number in the closed interval [Long.MIN_VALUE,Long.MAX_VALUE] (including Long.MIN_VALUE and Long.MAX_VALUE).


raw

public double raw()
Returns a 32 bit uniformly distributed random number in the open unit interval (0.0,1.0) (excluding 0.0 and 1.0).

Specified by:
raw in class RandomElement
Returns:
a random double in the range (0,1).
See Also:
RandomJava