correlations and coordinates

Harvey DAVIES (
Fri, 8 Aug 1997 10:55:14 +1000 (EST)

On Thu, 7 Aug 1997, John Caron wrote:

> I guess I don't understand what you mean by "coordinate system".

I assume you refer to my comment on following from John Sheldon:

> > >       Not applicable..."correlation" does not possess "coordinates" the
> > >       way we think of them.  [rest deleted]

My comment was:

> > I just want to emphasise my agreement by pointing out that one could
> > equally well be considering correlations between different variables at
> > the same station.  

I was merely pointing out that the variable which identifies the variables
being correlated (e.g. station, met. measurements) is essentially nominal in
nature.  It could be a coordinate variable with a string value (e.g. station
name, type of measurement (e.g. "annual precip.")).  I (perhaps wrongly) got
the feeling that some people were trying to re-cast such essentially
non-geographic cases into geographic frameworks by adding essentially
irrelevant (as far as calculating correlations) geographic hooks in the form
of lat/lon coordinate variables.


Harvey Davies, CSIRO Mathematical and Information Sciences,
723 Swanston Street, Carlton, Victoria 3053, Australia            
Phone: +61 3 9282 2623 or +61 3 9239 4556
  Fax: +61 3 9282 2600