Abstract
Some form of random sampling is employed in all simulation studies, so, some possibly substantial sampling errors are inevitable. Therefore, a new paradigm emerged: it is not always necessary to resort to randomness to generate inputs. Then, novel sampling methods were derived from this paradigm like for example Refined Descriptive Sampling (RDS). In this paper, we propose a software component under Linux called getRDS which implements an RDS number generator of high quality using the RDS method. It was highly tested by statistical tests and compared to the well known Mersenne Twister random number generator MT19937 (MT). We noticed that getRDS has passed better all tests than MT. Some illustrations of the uniformity are also given together with its comparison with MT through an M/M/1 simulation system. The obtained results through simulation demonstrate that the software component produces an accurate point estimates of the true parameters. Moreover, the getRDS random number generator can significantly improve the performance of the M/M/1 queues compared to MT since its variance reduction is over 50 % in some cases.
Similar content being viewed by others
Notes
Toolbox for pseudo and quasi random number generation and RNG tests available in http://cran.univ-lyon1.fr/web/packages/randtoolbox/
References
Chen X, Ankenman BE, Nelson BL (2012) The effects of common random numbers on stochastic kriging metamodels. ACM Trans Model Comput Simul 22(2):1–20
Matsumoto M, Nishimura T (2000) Dynamic creation of pseudorandom number generators. In: Niederreiter H, Spanier J (eds) Monte Carlo and Quasi Monte Carlo Methods, Springer, pp 56–69
Ourbih-Tari M, Aloui A (2009) Sampling methods and parallelism into Monte Carlo simulation. J Stat Adv Theory Appl 1(2):169–192
Ourbih-Tari M, Aloui A, Alioui A (2009) A software component which generates regular numbers from refined descriptive sampling. In: Proccedings of the European simulation modelling (ESM’2009) conference. Edited by Marwan Al-Akaidi. Leicester, United Kingdom, pp 23–25
Pidd M (2004) Computer simulation in management science, 5th edn. Wiley, Chichester
Ramberg JS, Schmeiser BW (1972) An approximate method for generating symmetric random variables. Commun ACM 15:987–990
Robert CP, Casella G (2004) Monte Carlo statistical methods, 2nd edn. Springer, New York
Saliby E (1990) Descriptive sampling: a better approach to Monte Carlo simulation. J Oper Res Soc 41(12):1133–1142
Saliby E (1997) Descriptive sampling: an improvement over latin hypercube sampling. In: Andradottir S, Healy KJ, Withers DH, Nelson BL (eds) Proceedings of winter simulation conference
Saliby E, Pacheco F (2002) An empirical evaluation of sampling methods in risk analysis simulation: Quasi Monte Carlo, descriptive sampling, and latin hypercube sampling. In: Yucesan E, Chen L, Snowdon J, Charnes JM (eds) Proceedings of the 2002 winter simulation conference, pp 1606–1610
Schellhorn H, Kidani F (2000) Variance reduction techniques for large scale risk management. In: Niederreiter H, Spanier J (eds) Monte Carlo and Quasi Monte Carlo Methods, Springer, pp 419–435
Stankiewicz R, Jajszczyk A (2010) Performance modeling of DiffServ meter/markers. Int J Commun Syst 23(12):1554–1580
Tari M, Dahmani A (2005a) The refining of descriptive sampling. Int J Appl Math Stat 3(M05):41–68
Tari M, Dahmani A (2005b) The three phase discrete event simulation using some sampling methods. Int J Appl Math Stat 3(D05):37–48
Tari M, Dahmani A (2006) Refined descriptive sampling: a better approach to Monte Carlo simulation. Simul Model Pract Theory 14:143–160
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Aloui, A., Zioui, A., Ourbih-Tari, M. et al. A general purpose module using refined descriptive sampling for installation in simulation systems. Comput Stat 30, 477–490 (2015). https://doi.org/10.1007/s00180-014-0545-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-014-0545-7