Random number generators and causality

We advance a prescription to randomize physical or algorithmic Random Number Generators (RNG's) that do not pass Marsaglia's DIEHARD test suite and discuss a special physical quantifier, based on an intensive statistical complexity measure, that is able to adequately assess the improvement...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Larrondo, H.A., Martín, M.T., González, C.M., Plastino, A., Rosso, O.A.
Formato: JOUR
Materias:
Random number generators
Statistical complexity
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_03759601_v352_n4-5_p421_Larrondo
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
Descripción
Sumario:We advance a prescription to randomize physical or algorithmic Random Number Generators (RNG's) that do not pass Marsaglia's DIEHARD test suite and discuss a special physical quantifier, based on an intensive statistical complexity measure, that is able to adequately assess the improvements produced thereby. Eight RNG's are evaluated and the associated results are compared to those obtained by recourse to Marsaglia's DIEHARD test suite. Our quantifier, which is evaluated using causality arguments, can forecast whether a given RNG will pass the above mentioned test. © 2005 Elsevier B.V. All rights reserved.

Ejemplares similares