Package jline.solvers.ldes.handlers

Class Solver_ssj.DiscreteUniformGen

java.lang.Object
umontreal.ssj.randvar.RandomVariateGen
jline.solvers.ldes.handlers.Solver_ssj.DiscreteUniformGen
Enclosing class:
Solver_ssj
public static class Solver_ssj.DiscreteUniformGen extends umontreal.ssj.randvar.RandomVariateGen
Random variate generator for DiscreteUniform, the uniform law on the w = max - min + 1 points min, min+1, ..., max.

The struct carries the law as the moment pair {mean, SCV} rather than as its bounds, exactly as it carries a Geometric, so the bounds are recovered here: var = SCV*mean^2 and var = (w^2-1)/12 give the integral width w = sqrt(12*var+1), and mean = (min+max)/2 then places it, which is exact and leaves a non-integral lower bound (a lattice shifted off the integers) where the user put it.

The draw is SSJ's UniformIntDist.inverseF, min + (int)(u*w), so it costs ONE uniform and the native C++ engine reproduces the same point.

  • Field Summary

    Fields inherited from class umontreal.ssj.randvar.RandomVariateGen

    dist, stream

  • Constructor Summary

    Constructors
    Constructor
    Description
    DiscreteUniformGen(umontreal.ssj.rng.RandomStream stream, double min, double width)
     

  • Method Summary

    Modifier and Type
    Method
    Description
    double
    nextDouble()
     

    Methods inherited from class umontreal.ssj.randvar.RandomVariateGen

    getDistribution, getStream, nextArrayOfDouble, nextArrayOfDouble, setStream, toString

    Methods inherited from class java.lang.Object

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

  • Constructor Details

    • DiscreteUniformGen

      public DiscreteUniformGen(umontreal.ssj.rng.RandomStream stream, double min, double width)

  • Method Details

    • nextDouble

      public double nextDouble()
      Overrides:
      nextDouble in class umontreal.ssj.randvar.RandomVariateGen