
Correlation 239
at orientation 0 and
φ
through u, shown in Figure 7.32. As Figure 7. 32 suggests, this summary
function can be estimated in the same way as the K-function.
The spatstat function pairorient computes an estimate of either the cumulative distribution
function or the probability density of angles. The default is to compute the probability density, using
a kernel smooth ing estimate:
> f <- pairorient(Y, r1=0.02, r2=0.05, sigma=5)
The result is an "fv" object. Although this could be plotted by plot.fv, the display is challeng ing
to interpret. For probability distributions of angles, it is usually better to display a rose diagram or
rose of ...