Research and Advances

Computer generation of gamma random variables—II

A rejection method is proposed for generating gamma variates with nonintegral shape parameter &agr;, &agr; > 1. This method is similar to other methods given by Fishman, Wallace, and Tadikamalla and is faster than these methods for &agr; > 2. The core storage requirements and the programming effort for the proposed method are similar to those of Wallace's or Tadikamalla's methods. The computational times for the proposed method remain fairly constant for medium and large values of &agr; and are superior to times obtained by Arhens and Dieter's method for all values of &agr;. The proposed method is simpler than Ahrens and Dieter's method.

Advertisement

Author Archives

Research and Advances

Computer generation of gamma random variables

A new method for generating random variables from the gamma distribution with nonintegral shape parameter &agr; is proposed. This method is similar to two other methods recently given by Wallace and Fishman. It is compared with Fishman's and Ahrens and Dieter's methods. The core storage requirements and programming effort for this method are similar to those of Fishman's method. The proposed method is the same as Fishman's method for 1 ≤ &agr; < 2 and is faster than Fishman's method for 3 ≤ &agr; ≤ 19. Also, the proposed method is much simpler than Ahrens and Dieter's method and is faster for &agr; ≤ 8.

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved