Re: Idempotence and "Replication Insensitivity" are equivalent ?

pamelafluente_at_libero.it wrote:
> I see what you mean.
>
> Hmmm...
> However, what one can observe is that for the same transformation this
> invariance property does not hold for the AVG, as well, and this does
> not seem to be a big problem ...

Whether or not it's a "big problem" depends entirely on your real life data model. By themselves, the statistics are meaningless numbers. *You* have to decide what makes sense in your model and whether the order statisctics invariance benefit is really useful.

>
> Both are invariant wrt linear transformations .... What do you think ?

I think that you may consider reading a book or two on statistics analysis if you are really interested in this kind of stuff.

>
> -P
Received on Mon Sep 25 2006 - 11:48:55 CDT