## AI-GEOSTATS: Re: setting up cokriging equations

Expand Messages
• Bettina I can t do math over the email but there are some papers Bill Harper and I wrote back in the mid-late 80s which you might find useful. Download from
Message 1 of 4 , Apr 19, 2004
• 0 Attachment
Bettina

I can't do math over the email but there are some
papers Bill Harper and I wrote back in the mid-late
http://uk.geocities.com/drisobelclark/resume
(click on Publications).

Basically, the principle is simple:

Consider each variable on each sample to be a
different 'data point'

Left hand side matrix is all semi-variograms between
each data point and each data point. Only, this time,
you have to remember which 'variable' is measured on
that datum and identify which semi-variogram model is
relevant.

Right hand side 'vector' is similar in that the
semi-variograms are between your estimated variable at
specified point and each data point. Again, you just
need to track which semi-variogram is relevant
(ordinary or cross).

The only real practical difference between co-kriging
and kriging is that the weights on your primary
variable (the one you are estimating) must sum to 1,
while the weights one each co-variable (individually)
must sum to zero to ensure unbiassedness. So, if, you
are using one co-variable you will have two extra
columns (and rows) on your left hand side matrix
to zero. If you have two co-variables, you will get
yet another row and column ensuring the weights on the
second co-variable add up to zero.

Does this help?
Isobel
http://geoecosse.bizland.com/whatsnew.htm

____________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping"

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org
• Hi Bettina I suggest the book An Introduction to Applied Geostatistics by Isaaks and Srivastava. Chapter 17 is devoted soley to cokriging. And as a bonus,
Message 2 of 4 , Apr 19, 2004
• 0 Attachment
Hi Bettina
I suggest the book "An Introduction to Applied Geostatistics" by Isaaks and
Srivastava. Chapter 17 is devoted soley to cokriging. And as a bonus,
Chapter 16 explains how to construct the linear model of coregionalization
for co-kriging.
Cheers, Ed

----- Original Message -----
From: "Bettina Berger" <Bettina_Berger@...>
To: <ai-geostats@...>
Sent: Monday, April 19, 2004 1:03 AM
Subject: AI-GEOSTATS: setting up cokriging equations

> Hello!
> Can anybody help me to set up the cokriging equations? I found the
> theoretical equations in various books (Cressie, Chiles, Wackernagel), but
I
> have a hard time to really understand them and to implement them into
> MATLAB. To be honest: I´m lost.
> I´m "only" a MechEng student, so it seems that my mathematical
possibilities
> are at its limits with kriging generally and furthermore with cokriging.
>
> What I would need are not only the general theoretical equations for
> Cokriging but also a practical example with real "numbers", which is
> calculated from beginning to the end, including finding the C-Matrix, D
> Matrix and the Kriging Weights.
> And: what exactly is the difference between kriging and co-kriging at all?
>
> Is there any article or book I can find that? Or is anyone of you able to
> write it down for me?
> Thank you so much,
> Cheers, Bettina
>
>
> --
> * To post a message to the list, send it to ai-geostats@...
> * As a general service to the users, please remember to post a summary of
any useful responses to your questions.
> * To unsubscribe, send an email to majordomo@... with no subject and
"unsubscribe ai-geostats" followed by "end" on the next line in the message
body. DO NOT SEND Subscribe/Unsubscribe requests to the list
> * Support to the list is provided at http://www.ai-geostats.org

--
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org
Your message has been successfully submitted and would be delivered to recipients shortly.