[R] attractive spatial point process?

James Marca jmarca at translab.its.uci.edu
Tue May 7 17:38:19 CEST 2002

Hi all,

before I roll my own, naive implementation of a particular 
spatial point process, I thought I'd ask whether this has 
already been done.

Specifically, I am looking for essentially the opposite of the 
Strauss and Mate'rn processes included in library(spatial) and 
documented in MASS.  I am examining daily travel patterns, focusing
on just the destinations of trips, not the movement itself.  As 
people tend to go to the same places repeatedly, I think an 
appropriate spatial point process would be one which has a 
large probability of visiting the neighborhood of a prior point
(but not the current point---this is a trip, after all), while at 
the same time having a non-zero probability of visiting a new point.  
Contrast this with the solid core/non-overlapping aspects of Strauss 
and SSI.  Perhaps my process should be called SSA? 

Following the implementation of Strauss, I was going to parameterize 
the size of the points and the repeat/new probabilities, write 
a little C program, then see how the process K value compares to 
my collected data.  

A moderate-level search of R resources didn't 
turn up such a process already, and a Google search was equally 
useless (although this is perhaps due to my cluelessness regarding 
what to call this kind of process).

Any pointers would be kindly appreciated, including "roll your 
own" and "read these refs."  

---thanks in advance,

r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch

More information about the R-help mailing list