B. Water <[email protected]> is trying to get Fishers exact
test to run on a 2x10 table with N = 1575.
>
> no. of obs 1575 2 var's: var1 had 2 categories and var2 had 10 categories.
>
> i tried:
>
> .tabulate var1 var2, exact
>
> and get this:
>
> "i've omitted the tabulation output"
>
> exceeded memory limits using exact(1); try again with larger #; see help tabulate for details
> r(911);
Clearly this is too large of a problem for -tabulate, exact- to handle. With
such a large N, I am sure the chi-squared p-values are valid. If you go
beyond exact(5), you've gone too far.
We made an update to -tabulate, exact- in version 9 (Jan. 20 2006) where we
gained some efficiency in the computation. We also added an enumeration log so
the user can see how big the problem is and whether a solution is feasible.
Judging by the size of N for your problem it still may not complete, but if you
want to send me your table I can see how it does.
-Rich
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
