[SciPy-Dev] binom_test incorrect behavior in the tails

Fernando Perez fperez.net@gmail....
Wed Mar 16 13:51:09 CDT 2011


On Wed, Mar 16, 2011 at 11:33 AM, Yaroslav Halchenko
<lists@onerussian.com> wrote:
>
> could anyone check if following behavior is still in place in the
> current master:
>
> $> python -c 'import scipy.stats,numpy; print numpy.array([scipy.stats.binom_test(a, 100) for a in [90, 98, 99, 100]])'
> [  1.53164509e-17   1.60428187e-19 1.66102449e-18   2.16868332e-09]
>

looking much better with master from today:

longs[scipy]> python -c 'import scipy.stats,numpy; print
numpy.array([scipy.stats.binom_test(a, 100) for a in [90, 98, 99,
100]])'
[  3.06329018e-17   7.96907286e-27   1.59349903e-28   1.57772181e-30]


cheers,

f


More information about the SciPy-Dev mailing list