Fisher's exact test for bigger than 2 by 2 contingency table

Question

Hi scipy stats has a implementation of Fisher's exact test but it is only for 2 by 2 contingency tables. I want to do the test on bigger than 2 by 2 tables. (5x2 ,5x3) I know there is fisher.test in R which can do the job but I want to do it in my python code

Anybody knows an python implementation of Fisher's exact test that can work on bigger tables?

Also I am not sure if it is ok to do Fisher's exact test on bigger than 2 by 2 tables.

Thanks

Answer 1

Yes, it is ok to do a Fisher's exact test on tables bigger than 2x2.

There currently aren't any clean, widely tested solutions out there in python. One solution would be to use rpy2 and call the R function from python:

import numpy as np
import rpy2.robjects.numpy2ri
from rpy2.robjects.packages import importr
rpy2.robjects.numpy2ri.activate()

stats = importr('stats')
m = np.array([[4,4],[4,5],[10,6]])
res = stats.fisher_test(m)
print 'p-value: {}'.format(res[0][0])
>> p-value: 0.668165917041

Another solution would be to dig into the C code that the R implementation uses and to call that code directly. Here is a link to someone's github project where they went back to the original fortran implementation and call that from python.