[Scipy-tickets] [SciPy] #1489: fisher_exact throws ValueErrors when row or col is 0, 0

SciPy Trac scipy-tickets@scipy....
Sun Aug 7 19:27:42 CDT 2011


#1489: fisher_exact throws ValueErrors when row or col is 0,0
--------------------------+-------------------------------------------------
 Reporter:  aihardin      |       Owner:  somebody   
     Type:  defect        |      Status:  new        
 Priority:  normal        |   Milestone:  Unscheduled
Component:  scipy.stats   |     Version:  0.9.0      
 Keywords:  fisher_exact  |  
--------------------------+-------------------------------------------------

Comment(by warren.weckesser):

 My experience with the Fisher exact test is limited to reading the Wolfram
 and wikipedia pages, so take this with the appropriate grain of salt:

 If there is a row or column of zeros, the p-value is 1.  The p-value is
 the sum of the probabilities of the observed data and "more extreme" data,
 conditioned on the marginal sums.  If either a row or column is all zero,
 then that data set is the only possible data set with the same row sums
 and column sums, so its probability is 1; there are no other possible data
 configurations to consider.

 The odds ratio formula will result in 0/0.  Perhaps NaN is the correct
 value for the odds ratio in this case.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1489#comment:5>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list