[Scipy-tickets] [SciPy] #1394: One-sided Fisher's Exact Test

SciPy Trac scipy-tickets@scipy....
Mon Feb 28 14:10:26 CST 2011


#1394: One-sided Fisher's Exact Test
-------------------------+--------------------------------------------------
 Reporter:  gotgenes     |       Owner:  somebody
     Type:  enhancement  |      Status:  new     
 Priority:  normal       |   Milestone:          
Component:  scipy.stats  |     Version:  0.8.0   
 Keywords:               |  
-------------------------+--------------------------------------------------
 Ticket #956 saw the introduction of two-sided Fisher's Exact Test (FET)
 into !SciPy in milestone:0.9.0, which is great; well done! I depend on
 one-sided FET, however.
 [http://stat.ethz.ch/R-manual/R-patched/library/stats/html/fisher.test.html
 R's `fisher.test` function] lets the user specify this in the
 `alternative` keyword argument, which can be either `'two.sided'`,
 `'greater'`, or `'less'`

 I have tried to understand the code, however, I am not a statistician, and
 there are absolutely no explanations to the algorithm implemented by
 `fisher_exact`. Code comments, links or docstrings on the algorithm, etc.
 would be very helpful and welcome.

 Perhaps all that is necessary for the one-sided test p-value is a call to
 `hypergeom.sf` (or `hypergeom.cdf`). I am not sure what to do for the odds
 ratio.

 By the way, the version for this ticket should be 0.9.0, however, that
 version number was unavailable at the time I filed this ticket.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1394>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list