Pseudorandom Number Generation
Chapter 8 reveals that every algorithm that generates a sequence of i.i.d. random samples from a probability distribution as output requires a sequence of i.i.d. random samples from u(0, 1) as input. To meet this need, every discrete-event simulation programming language provides a pseudorandom number generator that produces a sequence of nonnegative integers Z 1, Z 2,... with integer upper bound M > Z i ∀i and then uses U 1, U 2,..., where U i := Z i /M, as an approximation to an i.i.d. sequence from u(0, 1).
KeywordsRandom Number Generator Discretization Error Pseudorandom Number Combine Generator Primitive Root
Unable to display preview. Download preview PDF.
- Jannson, B. (1966). Random Number Generators, Almquist and Wiksell, Stockholm.Google Scholar
- L’Ecuyer, P., and T.H. Andres (1997a). A random number generator based on the combination of four LCGs, version: November 25, 1997, University of Montreal, available at http://www.iro.umontreal.ca/—lecuyer.Google Scholar
- L’Ecuyer, P. (1999). Personal communication.Google Scholar
- Overstreet, Jr., C.L. (1974). The relationship between the multiplicative and mixed generators modulo 2“, Computer Science Department, Virginia Polytechnic Institute and State University, Blacksburg.Google Scholar