[R] Exact Tests for RxC contingency tables

[Kurt Hornik]

>>>>> Mike Weale writes:

> I'm trying to perform exact tests on tables of approx. size 20x20,
> with about half the cells empty.  fisher.test in the ctest package
> appears to fall down with tables bigger than about 6x6.  Are there any
> other algorithms for exact tests implemented in other R-packages,
> which work on larger tables?

You can increase the workspace for the network algorithm in fisher.test
using argument `workspace'.  But I would think that many empty cells
will give it a real hard time.  I would go for the hybrid version or the
bootstrapped version of the chi-squared.test via

	chisq.test(x, simulate.p.value = TRUE, B = 2000)

-k




-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._