I've updated the Ziggurat normal and exponential
generator in SuppDists in accordance with
Leong, et.al (2005). A comment on the implementation
of the Ziggurat Method. Jour. Stat. Sci. 12-7, 1-4.
The fix takes care of problems for very large sets
of random values. I tested it a bit on small samples
and it seems ok. It will be on CRAN shortly.
Bob Wheeler --- http://www.bobwheeler.com/ ECHIP, Inc. --- Randomness comes in bunches.