[SciPy-User] scipy.stats.nanstd, bias and ddof

Keith Goodman kwgoodman@gmail....
Fri Jan 15 12:56:10 CST 2010


On Fri, Jan 15, 2010 at 10:48 AM,  <josef.pktd@gmail.com> wrote:
> Or maybe numpy should get a nanmean and nanvar, nanstd, similar to
> nansum ? Then it would be easier to depreciate like the other stats
> functions that moved to numpy.

That's a great idea. Adding nanstd to numpy would not break any code.

It would also be nice to have a nanmedian in numpy, one that doesn't
do a full sort. A pony would be nice too.


More information about the SciPy-User mailing list