chull(x, y, peel=F, maxpeel=<<see below>>, onbdy=peel, tol=.0001)

x,y:

coordinates of points. The coordinates can be given by two arguments that are vectors or by a single argument x which is a univariate time series, a complex vector, a matrix with 2 columns, or a list containing components named x and y. Missing values are not accepted.

peel:

logical flag: should successive convex hulls be peeled from the remaining points, generating a nested set of hulls?

maxpeel:

maximum number of hulls that should be peeled from the data. The default, if peel is TRUE, is to peel until all points are assigned to a hull. If maxpeel is given, peel is coerced to TRUE.

onbdy:

should points on the boundary of a convex hull (but not vertices of the hull) be included in the hull?

tol:

relative tolerance for determining if a third point is sufficiently collinear to two points in the hull to also be included in the hull.

If peel is TRUE, a list with components depth, hull and count.

depth:

a vector which assigns a depth to each point, i.e., the number of hulls surrounding the point (the outermost points are at depth 1). Outliers will have small values of depth, interior points relatively large values.

hull:

vector giving indices of the points on the hull. Along with count, determines all of the hull peels. The first count[1] values of hull determine the outermost hull, the next count[2] are the second hull, etc.

count:

counts of the number of points on successive hulls.

hull <- chull(x,y)
plot(x,y)
polygon(x[hull],y[hull],density=0)  # draw hull

p <- chull(corn.rain, corn.yield, peel=T)   # all hulls
which <- rep(seq(p$count), p$count)   # which peel for each pt
s <- split(p$hull, which)
plot(corn.rain, corn.yield)
for(i in seq(s)) {     # plot all peels
      j <- s[[i]]    # indices of points on ith peel
if (length(j)>2)
      polygon(corn.rain[j], corn.yield[j], density=0,lty=i)
}