[R-sig-Geo] Correlation between O&D

Thu Oct 13 14:40:40 CEST 2011

I'm still not entirely sure what you're getting at:  spatial correlation
of origin/destination locations to each other?  Or some characteristic
of the origin/destination pairs?  It sounds like the latter, but then
the question is "correlated with what"?

The choice of method is ultimately driven by the conceptual model of the
phenomena, which is what those questions are getting at.  If you can
answer those questions with some precision, finding an analytic method
that can implement the resulting model is fairly straightforward.

If you're still trying to sharpen your thinking (or even if you think
you have the abstractions nailed down and are just looking for
corresponding methods), I'd recommend starting with the "spatstat"
package (e.g. via the website www.spatstat.org) and following its
reference list into the world of spatial statistics.

If you're actually interested in stolen cars, you might take a look at
CrimeStat (http://www.icpsr.umich.edu/CrimeStat/) which rolls up
implementations of a lot of spatial statistics techniques that could be
applicable in tracking crimes and finding criminals.  The software is
"free as in beer" but it is not implemented in R (though I suppose one
might be able to wrap the provided library functions).  The various
analytic methods are mostly standard and reasonably well documented, so
any given analysis could be quite readily reproduced in the statistical
system of your choice.

I hope that helps!

Jeremy Raw, P.E.

FHWA Office of Planning

jeremy.raw at dot.gov

(202) 366-0986

From: Rich Heimann [mailto:heimann.richard at gmail.com] 
Sent: Wednesday, October 12, 2011 7:26 PM
To: Raw, Jeremy
Cc: r-sig-geo at r-project.org
Subject: Re: [R-sig-Geo] Correlation between O&D

Great! I hope you can tell me more. 

I am considering origin and destination to be linked data... Such as a
stolen car and a recovered car - rather than an independent point
pattern. I hope that clears things up? 

Regards,
Rich

Rich Heimann

 Tel: (571) 403-0119 
heimann.richard at gmail.com 

On Wed, Oct 12, 2011 at 3:16 PM, <jeremy.raw at dot.gov> wrote:

It all depends on what you mean by "origin and destination data".
Assuming I am allowed to make up my own definitions, the answer is
definitely "yes" :-)

My advice is to be somewhat more specific about the problem domain.

Jeremy Raw, P.E.
FHWA Office of Planning
jeremy.raw at dot.gov
(202) 366-0986 <tel:%28202%29%20366-0986> 

-----Original Message-----
From: r-sig-geo-bounces at r-project.org
[mailto:r-sig-geo-bounces at r-project.org] On Behalf Of Rich Heimann
Sent: Wednesday, October 12, 2011 2:24 PM
To: r-sig-geo at r-project.org
Subject: [R-sig-Geo] Correlation between O&D

Hello all,

I want to know if the list is aware of a way to determine whether origin
and
destination data is correlated - rather than looking at each as
independent?
Any advise?

Thanks,
Rich

Rich Heimann

 Richard Heimann | ITT Corporation | T: 571.403.0119 |
richard.heimann at itt.com
Richard Heimann | University of Maryland, Baltimore County | T:
571.403.0119
| rheimann at umbc.edu

Contact Me <http://www.linkedin.com/in/rheimann>
LinkedIn<http://www.linkedin.com/in/rheimann>
<https://twitter.com/rheimann> Twitter <https://twitter.com/rheimann>
<https://plus.google.com/117471169530873553158> Google
Plus<https://plus.google.com/117471169530873553158>

Chat [image: Google Talk] heimann.richard [image: Skype] heimann.rich

 [image: Twitter] <http://twitter.com/rheimann> Latest tweet: An

Intuitive
Explanation of Bayes' Theorem: http://t.co/6DAh8O43

Follow @rheimann <http://twitter.com/rheimann> Reply
<http://twitter.com/?status=@rheimann%20&in_reply_to_status_id=123943215
412953090&in_reply_to=rheimann
<http://twitter.com/?status=@rheimann%20&in_reply_to_status_id=123943215
%0A412953090&in_reply_to=rheimann> >
Retweet
<http://twitter.com/?status=RT%20%40rheimann%3A%20An%20Intuitive%20Expla
nation%20of%20Bayes%27%20Theorem%3A%20http%3A%2F%2Ft.co%2F6DAh8O43>
 22:08 Oct-11<http://twitter.com/rheimann/statuses/123943215412953089>
 Get this email app!
<http://www.wisestamp.com/apps/twitter?utm_source=extension&utm_medium=e
mail&utm_term=twitter&utm_campaign=apps
<http://www.wisestamp.com/apps/twitter?utm_source=extension&utm_medium=e
%0Amail&utm_term=twitter&utm_campaign=apps> >

 Please consider your environmental responsibility. Before printing this
e-mail message, ask yourself whether you really need a hard copy.

       [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20111013/2c67db4b/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 332 bytes
Desc: image001.jpg
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20111013/2c67db4b/attachment.jpe>