Re: Idempotence and "Replication Insensitivity" are equivalent ?

vc ha scritto:

> Missed that:
>
> pamelafluente_at_libero.it wrote:
> [...]
> > For a continuous cdf the median is a
> > point.
>
> Obviously, that is not always the case. With a continuous probability
> distribution, the median of the pdf also may not be unique.

In theory its right. I meant for a cdf with anywhere positive density, cases of continuous cdf's with density equal to 0 around the median are actually never seen. Did you ever see them? Hmm, perhaps some mixtures ...

-P Received on Tue Sep 26 2006 - 09:45:47 CDT