Leonardo Music Journal, Vol. 1, No. 1, 1991, pp. 55-70.
This article surveys methods for programming computers to realize statistical distributions, strictly isolating distributions per se from random generators so that the same methods may serve random, determinate and chaotic applications. Considerations in choosing appropriate distributions are examined. Numerical methods are provided to realize arbitrary distributions of both discrete and continuous types. Many standard distributions are motivated using probabilistic models. Distributions are illustrated by graphs showing both ideal densities and actual random populations.