Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Fischer Exact Test again


From   Richard Goldstein <richgold@ix.netcom.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Fischer Exact Test again
Date   Tue, 11 Jun 2002 08:20:44 -0400

While there are fast algorithms for Fisher's (and other exact tests --
see, e.g., StatXact), Stata does not use a fast algorithm; you will
eventually get a result, but it will be faster to buy StatXact

Rich Goldstein

JiTian Sheu wrote:
> 
> Dear listers:
> I sent this subject yesterday(or the day before yesterday), and got the
> reply from Mr. Nick.
> I do appreciate your assistence on this subject.
> And I did try to use Auto dataset and see whether for a 2XC(C>2) table, I
> can obtain Fischer exact p-value or not.
> And It does work.
> 
> However, the "auto" dataset is small, the observation is approximately only
> 70. Hence, STATA takes few seconds to calculate the Fischer Exact p-value
> for this 2X5 table(foreign vs rep78).
> My problem is that my dataset is not so big(the number of observations for
> comparison is approximately 2000). This moring, I use -tabulate- to obtain
> the Fischer exact p-value. However, up to now, I still can not get the
> results.
> (This means that for a computer with 312M memory, STATA needs at least 10
> hours for calculating Fischer Exact p-value.)
> I am just wondering whether there is any lister having this kind of
> situation before or not, and how did you solve this situation?
> 
> Up to the moment I send this email, I still can not get the result yet.
> Thank you very much for any of you who can respond this email. Thanks again.
> 
> JT
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index