Re: [S] CoKriging Software

Anthony Rossini (rossini@math.sc.edu)
02 Apr 1998 10:15:23 -0500


We did this a number of years ago with Geoeas (which has a cokriging
module). We were less then happy with the results of our analysis
(not the program, but it's yet another interface). Similar setup, but
using completely different colocated variables - ozone, traffic data,
population density, a few others, for a metropolitan region.

It's not clear that cokriging's assumptions are reasonably met by air
pollution data, but that's for you to determine.

I would be interested in finding out if this is possible in Splus, as
well, without simply doing the geoeas/splus interface.

(I would have responded privately, except for the silly munged headers
which are completely annoying -- if you don't want spam, get your
sysadmin to block it at the site level, not by being silly -- a good
spammer can just multiply his list by 1000 (i.e. for each
email address, estimate the 1000 possible reasonable estimates
which might be the real one, and select the 50 or so which might be
valid -- a nice statistical estimation problem)).

See for example, the MAPS project at http://www.vix.com/

best,
-tony

-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news