I came across a p-value greater than one in a Fisher exact's test. The output was the following Fisher's exact test data: rbind(c(1, 1), c(71, 77)) p-value = 1.0001 alternative hypothesis: two.sided
Strangely enough, this is what I got: Fisher's exact test data: rbind(c(1, 1), c(71, 77)) p-value = 1 alternative hypothesis: two.sided -- Cande V. Ananth, PhD, MPH ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Not only the previous strange p-value, but also the other one by using S-PLUS 4.0 Fisher's exact test data: rbind(c(1, 1), c(91, 98)) p-value = 1.0001 alternative hypothesis: two.sided Yen-Hong Kuo B
Once again, it works fine using S-Plus ver 3.4, operating on SunOS 5.3 (Sun SPARC). The bug, I believe, is is in S-Plus 4.0. fisher.test(rbind(c(1, 1), c(91, 98))) Fisher's exact test data: rbind(c(
The Fisher exact test problem also comes out wrong in S+ 3.3 for Windows, so EVEN I have to admit (although I hated 4.0 so much I reverted) that it is not a Version 4 problem. fisher.test(rbind(c(1,
Very strange enough I get the same result als marcia: fisher.test(rbind(c(1, 1), c(71, 77))) Fisher's exact test data: rbind(c(1, 1), c(71, 77)) p-value = 1.0001 alternative hypothesis: two.sided Thi
I used to work at software testing (for one of the stat software company). And all programs have there bugs, with some strange data sets, on some systems. Any software is system dependent up to some
According to Arnold.Dekkers@rivm.nl: Could this be another undocumented Pentium bug? Has anyone tried it using a good old 486 processor, or any of the Cyrix, IBM or AMD processors? Patrick -- This me
Not a Pentium problem. These were run in Splus 3.4 on an SGI with an Irix operating system: Fisher's exact test data: matrix(c(1, 1, 70, 70), nc = 2) p-value = 1 alternative hypothesis: two.sided Fis
... rest snipped .... Doesn't occur for me with 4.0r3 on a Pentium Pro NT4sp3 machine: Fisher's exact test data: matrix(c(1, 1, 71, 71), nc = 2) p-value = 1 alternative hypothesis: two.sided Strange