Generation of discrete random variables on vector computers for Monte Carlo simulations

Simple item page
dc.contributor.authorSarno, R.
dc.contributor.authorBhavsar, V., C.
dc.contributor.authorHussein, E., M., A.
dc.date.accessioned2023-03-01T18:28:36Z
dc.date.available2023-03-01T18:28:36Z
dc.date.issued1990
dc.description.abstractThe paper reviews existing methods for generating discrete random variables and their suitability for vector processing. A new method for generating discrete random variables for use in vectorized Monte Carlo simulations is presented. The method uses the concept of importance sampling and generates random variables by employing uniform distribution to speedup the computation. The sampled random variables are subsequently adjusted so that unbiased estimates are obtained. The method preserves both mean and variance of the original distribution. It is demonstrated that the method requires simpler coding and shorter execution time for both scalar and vector processing, when compared with other existing methods. The vectorization speedup of the method is demonstrated on an IBM 3090-180 machine with a vector facility. Keywords: discrete random variables, importance sampling, Monte Carlo simulation, parallel processing, supercomputing, vector processing
dc.description.copyrightCopyright @ R. Sarno, V. C. Bhavsar, and E. M. A. Hussein, 1990.
dc.identifier.urihttps://unbscholar.lib.unb.ca/handle/1882/14860
dc.rightshttp://purl.org/coar/access_right/c_abf2
dc.subject.disciplineComputer Science
dc.titleGeneration of discrete random variables on vector computers for Monte Carlo simulations
dc.typetechnical report

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
item.pdf
Size:
603.58 KB
Format:
Adobe Portable Document Format
Download

Collections

Research Reports