Abstract
Random number generators (RNGs) are commonly used in simulations. The overlapping serial test is an important test that examines the randomness of RNGs. Its theory has been well-developed but its true ability for rejecting poor RNGs is not known. In this paper, we study the power of the test against the RNGs included in the widely spread GNU Scientific Library. By systematically varying the choices of parameters of the test, we find a fine-tuned version that rejects 29 RNGs out of the total of 57 in the library. We will like to warn users not to use these 29 RNGs.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Good, I.J.: The serial test for sampling numbers and other tests for randomness. In: Proceedings of Cambridge Philosophical Society, vol. 49, pp. 276–284 (1953)
Knuth, D.: The Art of Computer Programming, 3rd edn. vol. 2. Addison-Wesley, Reading (1998)
Marsaglia, G.: A current view of random number generators. In: Computer Science and Statistics: Proceedings of the 16th Symposium on the Interface, Atlanta, Georgia, pp. 151–158 (1984)
Marsaglia, G.: Monkeying with the Goodness-of-Fit Test. Journal of Statistical Software (2005)
Pearson, K.: On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. Philosophical Magazine 50(5), 157–175 (1900)
Tsang, W., Pang, C.Y.: The Mathematics of the Overlapping Chi-square Test. Technical Report TR-2006-12, Department of Computer Science, The University of Hong Kong (September 2006)
Marsaglia, G., Tsang, W.: Some difficult-to-pass tests of randomness. Journal of Statistical Software 7(3), 1–8 (2002)
Anderson, T., Darling, D.: Asymptotic Theory of Certain Goodness of Fit Criteria Based on Stochastic Processes. The Annals of Mathematical Statistics 23(2), 193–212 (1952)
Marsaglia, G., Marsaglia, J.: Evaluating the Anderson-Darling distribution. Journal of Statistical Software 9(2), 1–5 (2004)
Marsaglia, G., Zaman, A.: Monkey tests for random number generators. Computers and Mathematics with Applications 26(9), 1–10 (1993)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Xu, X., Tsang, W.W. (2007). An Empirical Study on the Power of the Overlapping Serial Test. In: Park, JW., Kim, T.G., Kim, YB. (eds) AsiaSim 2007. AsiaSim 2007. Communications in Computer and Information Science, vol 5. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-77600-0_32
Download citation
DOI: https://doi.org/10.1007/978-3-540-77600-0_32
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-77599-7
Online ISBN: 978-3-540-77600-0
eBook Packages: Computer ScienceComputer Science (R0)