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

Re: st: Fischer Exact Test again

From   Jeph Herrin <>
Subject   Re: st: Fischer Exact Test again
Date   Mon, 10 Jun 2002 22:36:35 -0400

The long wait is not too surprising - you should skip Fisher's unless you
have small numbers (<5?) in some cells. If with 2000 observations you have
small numbers in some cells, you should combine some categories - or
be prepared to wait a day or two.


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:
> *
> *
> *

Jeph Herrin, Ph.D.

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index