On Sat, Aug 28, 2021, 8:34 AM Stephen J. Turnbull <stephenjturnbull@gmail.com> wrote:

David Mertz, Ph.D. writes:
> > NANs do not necessarily represent missing data.

> I think in the context of `stats` they do. But this is color of bikeshed, and I defer to you, of course.

I have a distribution for you: Cauchy. :-)

Oh? Because NaN is the *result* of `stats.variance(cauchy)`?

It still seems like as INPUTS to a stats function NaN ~= missing.