[R] p-value > 1 in fisher.test()

[Don MacQueen]

At 10:01 PM +0200 6/3/05, Peter Dalgaard wrote:
>"Yi-Xiong Zhou" <yzhou at sdsc.edu> writes:
>
>>  The following contingency table generates p-value > 1 from fisher.test()
>>
>> 
>>
>>  ff = c(0,10,250,5000); dim(ff) = c(2,2); fhisher.test(ff)$p.value
>
>On this system (PIII, Fedora Core 3):
>
>  > ff = c(0,10,250,5000); dim(ff) = c(2,2); fisher.test(ff)$p.value -1
>[1] 1.384892e-12
>
>Is that interesting?

And on this system (powerpc, darwin7.8.0)

>  ff = c(0,10,250,5000); dim(ff) = c(2,2); fisher.test(ff)$p.value -1
[1] -2.603184e-11

About equally interesting.

-- 
--------------------------------------
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA