Inverting the symmetrical beta distribution

ABSTRACT

We propose a fast algorithm for computing the inverse symmetrical beta distribution. Four series (two around x = 0 and two around x = 1/2) are used to approximate the distribution function, and its inverse is found via Newton's method. This algorithm can be used to generate beta random variates by inversion and is much faster than currently available general inversion methods for the beta distribution. It turns out to be very useful for generating gamma processes efficiently via bridge sampling.

REFERENCES

• 1: Abramowitz, M. and Stegun, I. A. 1970. Handbook of Mathematical Functions. Dover, New York.
• 2: Avramidis, T. and L'Ecuyer, P. 2006. Efficient Monte Carlo and Quasi-Monte Carlo option pricing with the variance-gamma model. Manage. Sci. to appear.
• 3: Brown, B. W., Lovato, J., and Russell, K. 1994. Library of C routines for cumulative distribution functions, inverses, and other parameters. http://odin.mdacc.tmc.edu/anonftp/#DCDFLIB.
• 4: Caflisch, R. E. and Moskowitz, B. 1995. Modified Monte Carlo methods using quasi-random sequences. In Monte Carlo and Quasi-Monte Carlo Methods in Scientific Computing, H. Niederreiter and P. J.-S. Shiue, eds. Lecture Notes in Statistics 106. Springer Verlag, New York, 1--16.
• 5: Devroye, L. 1986. Non-Uniform Random Variate Generation. Springer Verlag, New York.
• 6: Armido R. Didonato , Alfred H. Morris, Jr., Algorithm 708; significant digit computation of the incomplete beta function ratios, ACM Transactions on Mathematical Software (TOMS), v.18 n.3, p.360-373, Sept. 1992 [doi>10.1145/131766.131776]
• 7: Fox, B. L. 1999. Strategies for Quasi-Monte Carlo. Kluwer Academic, Boston.
• 8: Walter Gautschi, Algorithm 222: Incomplete beta functions ratios, Communications of the ACM, v.7 n.3, p.143, March 1964 [doi>10.1145/363958.363968]
• 9: Walter Gautschi, Certification of algorithm 222: Incomplete beta function ratios, Communications of the ACM, v.7 n.4, p.244, April 1964 [doi>10.1145/364005.364073]
• 10: Hogg, R. V. and Craig, A. F. 1995. Introduction to Mathematical Statistics, 5th ed. Prentice-Hall.
• 11: Hörmann, W., Leydold, J., and Derflinger, G. 2004. Automatic Nonuniform Random Variate Generation. Springer Verlag, Berlin.
• 12: Averill M. Law , W. David Kelton, Simulation Modeling and Analysis, McGraw-Hill Higher Education, 1997
• 13: Pierre L'Ecuyer, Quasi-Monte Carlo methods in finance, Proceedings of the 36th conference on Winter simulation, December 05-08, 2004, Washington, D.C.
• 14: Moshier, S. L. 2000. Cephes math library. http://www.moshier.net.
• 15: Moskowitz, B. and Caflisch, R. E. 1996. Smoothness and dimension reduction in Quasi-Monte Carlo methods. J. Math. Comput. Modeling 23, 37--54.
• 16: Peizer, D. B. and Pratt, J. W. 1968. A normal approximation for binomial, F, beta, and other common related tail probabilities. J. American Statistical Association 63, 1416--1456.
• 17: Jerome Spanier , Keith B. Oldham, An atlas of functions, Taylor & Francis/Hemisphere, Bristol, PA, 1987
• 18: Ulrich, G. 1984. Computer generation of distributions on the m-sphere. App. Statistics 33, 158--163.

INDEX TERMS

Primary Classification:
  G. Mathematics of Computing
  G.4 MATHEMATICAL SOFTWARE
      Subjects: Algorithm design and analysis

Additional Classification:
  I. Computing Methodologies
  I.6 SIMULATION AND MODELING

General Terms:
Algorithms, Performance

Keywords:
Random variate generation, inversion method, quantiles, symmetrical beta distribution