Loading...

An algorithm for improving the behavior of strings of pseudo-random numbersfor variance reduction in sampling from gamma distribution

Moheghi, H ; Sharif University of Technology

467 Viewed
  1. Type of Document: Article
  2. Abstract:
  3. This research effort attempts to investigate the conditions under whichbetter variance reduction can be achieved in simulation of a system where randomdeviates are generated by the acceptance-rejection technique. To inducecorrelation among the generated deviates, both antithetic and stratifiedsampling will be considered to study the magnitude of variance reduction inestimating the renewal function for a single component replacement process withGamma distributed life. The negative exponential and the Cheng envelopingfunctions will be considered and all four combinations of correlationinduction-enveloping function will be tested. We will propose the correctapproach in incorporating streams of pseudo-random numbers in the variancereduction algorithms which use a single random number stream to guarantee theiruniform distribution as well as the correct correlation for various values ofGamma shape parameters. We consider a much wider spectrum of values for theGamma shape parameter and propose an algorithm on whose basis a higher degree ofvariance reduction can be attained. © 2009 IEEE
  4. Keywords:
  5. Variance reduction ; Acceptance-rejection technique ; Replacement process ; Single seed streams ; Uniform behavior ; Variance reductions ; Algorithms ; Exponential functions ; Industrial engineering ; Random number generation ; Parameter estimation
  6. Source: 2009 International Conference on Computers and Industrial Engineering, CIE 2009, 6 July 2009 through 9 July 2009, Troyes ; 2009 , Pages 1580-1585 ; 9781424441365 (ISBN)
  7. URL: http://ieeexplore.ieee.org/document/5223764