Visit each point in a point pattern, find the neighbouring points, and compile a frequency table of the marks of these neighbour points.
marktable(X, R, N, exclude=TRUE, collapse=FALSE)A marked point pattern.
An object of class "ppp".
Neighbourhood radius. Incompatible with N.
Number of neighbours of each point. Incompatible with R.
Logical. If exclude=TRUE, the neighbours of a point
do not include the point itself. If exclude=FALSE,
a point belongs to its own neighbourhood.
Logical. If collapse=FALSE (the default) the results for
each point are returned as separate rows of a table.
If collapse=TRUE, the results are aggregated according to the
type of point.
A contingency table (object of class "table").
If collapse=FALSE, the table has one row for
each point in X, and one column for each possible mark value.
If collapse=TRUE, the table has one row and one column
for each possible mark value.
This algorithm visits each point in the point pattern X,
inspects all the neighbouring points within a radius R of the current
point (or the N nearest neighbours of the current point),
and compiles a frequency table of the marks attached to the
neighbours.
The dataset X must be a multitype point pattern, that is,
marks(X) must be a factor.
If collapse=FALSE (the default),
the result is a two-dimensional contingency table with one row for
each point in the pattern, and one column for each possible mark
value. The [i,j] entry in the table gives the number of
neighbours of point i that have mark j.
If collapse=TRUE, this contingency table is aggregated
according to the type of point, so that the result is a contingency
table with one row and one column for each possible mark value.
The [i,j] entry in the table gives the number of
neighbours of a point with mark i that have mark j.
To perform more complicated calculations on the neighbours of every
point, use markstat or applynbd.
# NOT RUN {
head(marktable(amacrine, 0.1))
head(marktable(amacrine, 0.1, exclude=FALSE))
marktable(amacrine, N=1, collapse=TRUE)
# }
Run the code above in your browser using DataLab