[SciPy-User] int32 overflow and sqrt of -ve number in scipy.stats.wilcoxon

Simon Baldwin simonb@google....
Fri Mar 22 10:57:35 CDT 2013


 <josef.pktd <at> gmail.com> writes:
> 
> On Fri, Mar 22, 2013 at 8:43 AM, Simon Baldwin <simonb <at> google.com> wrote:
> > I have run into two distinct but possibly connected failure modes in
> > scipy.stats.wilcoxon.  Before bug-reporting, I'd appreciate any
> > thoughts on these.
> ...
> 
> this sounds like the same problem that rankdata had and Warren recently fixed.
> 
> My guess is that we can implement wilcoxon using the new rankdata
> (maybe, if we can get the tie counts).
> Otherwise, it would need a change to the c code.
> 
> (old code that wasn't designed for modern computers ?)

Thanks for the note.  The offending computation is

  corr += 0.5*si*(si*si-1.0)

so I think just converting si from int32 to int64 or float before use would 
solve the majority of cases.


> 
> ...
> 
> That might be a numerical precision issue in a corner case (-0) that should be
> explicitly handled.

I think it may be a more egregious error than that.  In this example the value 
being passed to sqrt is -443.05555555555554.

So far I haven't found an easy workround, not least because I don't yet 
understand the ties loop in scipy.stats.wilcoxon.


> 
> Please open a ticket.

Will do.  Thanks.

--
Google UK Limited | Registered Office: Belgrave House, 76 Buckingham
Palace Road, London SW1W 9TQ | Registered in England Number: 3977902




More information about the SciPy-User mailing list