Re: A new proof of the superiority of set oriented approaches: numerical/time serie linear interpolation

On Apr 30, 1:26 pm, "David Cressey" <cresse..._at_verizon.net> wrote: [Snipped]
> > You have not told me what you thought about the idea of using
> > interpolation as a possible computing method to systematically handle
> > missing data.
>
> You are right. I'm still thinking about that. Off the top of my head, I'd
> suggest that
> interpolation is useful where data points represent a finite sample of some
> kind of continuum, and that some, but not all, situations of missing data
> lendthemselves to that description. If this thinking sounds incomplete,
> that's because it is.
Precisely. Which is why temporal data and series perfectly meet the criteria of representing a continuum. Received on Mon Apr 30 2007 - 14:40:06 CEST