[R] Algorithms for coincidences

[Thomas Lumley]

On Tue, 6 Nov 2007, Greg Snow wrote:

> A paper that may help you:
>
> "Methods for Studying Coincidences", Persi Diaconis; Frederick
> Mosteller.  Journal of the American Statistical Association, vol 84, no.
> 408 (Dec., 1989), 853-861.

The pbirthday()/qbirthday() functions use the approximations in this paper 
and the help page references it, so it may not be much further assistance.

If the problem is parameter values that the functions don't handle, rather 
than insufficient precision, I would suggest simulation.

 	-thomas

Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle