Analysing spatial point patterns in R - CSIRO
Analysing spatial point patterns in R - CSIRO
Analysing spatial point patterns in R - CSIRO
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
118 Distance methods for <strong>po<strong>in</strong>t</strong> <strong>patterns</strong><br />
Thus, assum<strong>in</strong>g that the <strong>po<strong>in</strong>t</strong> process is homogeneous, we are able to compute an unbiased<br />
and reasonably accurate estimate of the empty space function F def<strong>in</strong>ed by (13).<br />
To <strong>in</strong>terpret this estimate, a useful benchmark is the Poisson process. Notice that d(u,X) > r<br />
if and only if there are no <strong>po<strong>in</strong>t</strong>s of X <strong>in</strong> the disc b(u,r) of radius r centred on u. For a<br />
homogeneous Poisson process of <strong>in</strong>tensity λ, the number of <strong>po<strong>in</strong>t</strong>s fall<strong>in</strong>g <strong>in</strong> b(u,r) is Poisson<br />
with mean µ = λarea(b(u,r)) = λπr 2 , so the probability that there are no <strong>po<strong>in</strong>t</strong>s <strong>in</strong> this region<br />
is exp(−µ) = exp(−λπr 2 ). Hence for a Poisson process<br />
F pois (r) = 1 − exp(−λπr 2 ). (16)<br />
Typically we compare ̂F(r) with the value of F pois (r) obta<strong>in</strong>ed by plugg<strong>in</strong>g <strong>in</strong> the estimated<br />
<strong>in</strong>tensity ˆλ = n(x)/area(W). Values ̂F(r) > F pois (r) suggest that empty space distances <strong>in</strong> the<br />
<strong>po<strong>in</strong>t</strong> pattern are shorter than for a Poisson process, suggest<strong>in</strong>g a regularly space pattern; while<br />
values ̂F(r) < F pois (r) suggest a clustered pattern.<br />
19.2.3 Implementation <strong>in</strong> spatstat<br />
The functionFest computes estimates of F(r) us<strong>in</strong>g several edge corrections, and the benchmark<br />
value for the Poisson process.<br />
> data(cells)<br />
> plot(cells)<br />
> Fc Fc<br />
Function value object (class fv)<br />
for the function r -> F(r)<br />
Entries:<br />
id label description<br />
-- ----- -----------<br />
r r distance argument r<br />
theo F[pois](r) theoretical Poisson F(r)<br />
cs F[cs](r) Chiu-Stoyan estimate of F(r)<br />
rs F[bord](r) border corrected estimate of F(r)<br />
km F[km](r) Kaplan-Meier estimate of F(r)<br />
hazard hazard(r) Kaplan-Meier estimate of hazard function lambda(r)<br />
--------------------------------------<br />
Default plot formula:<br />
. ~ r<br />
<br />
Recommended range of argument r: [0, 0.085]<br />
Available range of argument r: [0, 0.2975]<br />
Copyright<strong>CSIRO</strong> 2010