[R] Testing the R RNGs

B D McCullough bdm25 at drexel.edu
Fri Jun 13 05:33:52 CEST 2003


I have applied L'Ecuyer's TESTU01 suite of RNG tests 
to the RNGs in R.  TESTU01 offers three increasingly
more stringent suites, called "Small Crush", "Crush" and
"Big Crush".  If a particular RNG fails Small Crush, there
is no need to apply Big Crush.

Below I summarize the results:

                  Number of Tests Failed
		 Small Crush   Crush   Big Crush
Wichmann-Hill              1              3        NA
Multicarry                     1            13        NA
Super-Duper                1              9        NA
Mersenne                    --             --          --
TAOCP                      0              0          0
TAOCP-2002              0              0          0

NA: Not Applied because not necessary
--: couldn't duplicate RNG output, so test not run

A more detailed description, R-RNGTests.txt
will be available at via my homepage until the end
of June 2003: www.pages.drexel.edu/~bdm25
look for the link at the bottom of the page.

For reasons described in the R-RNGTests.txt, I
could not test the Mersenne RNG.  This will 
require someone with a better knowledge of
the C language than I possess.  If anyone can
figure it out, I would be most grateful.  Please
reply directly to me.

Regards,

Bruce McCullough




More information about the R-help mailing list