Abstract
The most efficint algorithms for sampling from the standard normal distribution require long lists of constants. The size of these tables grows with the employed precision. By adapting A.J. Walker's “alias method” to the normal distribution a sampling procedure is developed which needs only three fixed tables of 128 bytes each. The new method is as fast as its competitors and easier to implement.
Zusammenfassung
Die effizientesten Algorithmen für Stichproben von der Standardnormalverteilung benötigen lange Listen von Konstanten. Die Größe dieser Tafeln wächst mit der verwendeten Präzision. Durch eine Anpassung der “Aliasmethode” von A.J. Walker an die Normalverteilung wird eine Stichprobenprozedur entwicklet, die nur drei feste Tafeln von je 128 Bytes braucht. Die neue Methode ist ebenso schnell wie ihre Konkurrenten und leichter zu implementieren.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Ahrens, J. H.: How to avoid logarithms in comparisons with uniform random variables. Computing41, 163–166 (1988).
Ahrens, J. H., Dieter, U.: Computer methods for sampling from the exponential and normal distributions. Comm. ACM15, 873–882 (1972).
Ahrens, J. H., Dieter, U.: Extensions of Forsythe's method for random sampling from the normal distribution. Math. of Computation27, 927–937 (1973).
Ahrens, J. H., Dieter, U.: Efficient table-free sampling methods for the exponential, Cauchy and normal distributions. Comm. ACM31, 1330–1337 (1988).
Box, G. E. P., Muller, M. E.: A note on the generation of random noomal deviates. Ann. Math. Statistics29, 610–611 (1959).
Dervoye, L.: Non-Uniform Random Variate Generation. New York-Berlin-Heidelberg-Tokyo: Springer-Verlag 1986.
Fishman, G. S.: Principles of Discrete Event Simulation. New York: Wiley 1978.
Forsythe, G. E.: Von Neumann's comparison method for random sampling from the normal and other distributions. Math. of Computation26, 817–826 (1972).
Kinderman, A. J., Monahan, J. F.: Computer generation of random varaibles using the ratio of uniform deviates. ACM Trans. Math. Software3, 257–260 (1977).
Knuth, D. E.: The Art of Computer Programming, Vol. II: Seminumerical Algorithms. Reading, Mass.: Addison-Wesley 1969 and 1981.
Marsaglia, G.: Improving the polar method for generating a pair of normal random varaibles. Boeing Sci. Res. Labs. Seattle,. Washington, D1-82-0203, 1962.
Marsaglia, G.: The exact-approximation method for generating random varaibles in a computer. J. Amer. Statist. Assoc.79, 218–221 (1984).
Marsaglia, G., MacLaren, M. D., Bray, T. A.: A fast procedure for generating normal random varaibles. Comm. ACM7, 4–10 (1964).
Neumann, J. v.: Various techniques used in connectoin with random digits. Monte Carlo methods. National Bureau of Standards AMS12, 36–38 (1951).
Walker, A. J.: New fast method for generating discrete random numbers with arbitrary frequency distributions. Electron. Lett.10, 127–128 (1974).
Walker, A. J.: Fast generation of uniformly distributed pseudo random numbers with floating point representation. Electron. Lett.10, 553–554 (1974).
Walker, A. J.: An efficient method for generating discrete random varaibles with general distributions. ACM Trans. Math. Software3, 253–256 (1977).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Ahrens, J.H., Dieter, U. An alias method for sampling from the normal distribution. Computing 42, 159–170 (1989). https://doi.org/10.1007/BF02239745
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02239745