spdep (version 0.1-13)

knearneigh: K nearest neighbours for spatial weights

Description

The function returns a matrix with the indices of regions belonging to the set of the k nearest neighbours of each other. If lonlat = TRUE, Great Circle distances are used.

Usage

knearneigh(x, k=1, lonlat = FALSE)

Arguments

x

matrix of region point coordinates

k

number of nearest neighbours to be returned

lonlat

TRUE if point coordinates are longitude-latitude decimal degrees, in which case distances are measured in kilometers

Value

• A list of class knn
• nn
  integer matrix of region number ids
• np
  number of input points
• k
  input required k
• dimension
  number of columns of x
• x
  input coordinates

Details

The underlying C code is based on the knn function in the class package in the VR bundle.

See Also

knn, dnearneigh, knn2nb

Examples

data(columbus)
col.knn <- knearneigh(coords, k=4)
plotpolys(polys, bbs, border="grey")
plot(knn2nb(col.knn), coords, add=TRUE)
title(main="K nearest neighbours, k = 4")
data(state)
us48.fipsno <- read.geoda(system.file("etc/weights/us48.txt",
 package="spdep")[1])
m50.48 <- match(us48.fipsno$State.name, state.name)
xy <- as.matrix(as.data.frame(state.center))[m50.48,]
llk4.nb <- knn2nb(knearneigh(xy, k=4, lonlat=FALSE))
gck4.nb <- knn2nb(knearneigh(xy, k=4, lonlat=TRUE))
plot(llk4.nb, xy)
plot(diffnb(llk4.nb, gck4.nb), xy, add=TRUE, col="red", lty=2)
title(main="Differences between Euclidean and Great Circle k=4 neighbours")
summary(llk4.nb, xy, lonlat=TRUE)
summary(gck4.nb, xy, lonlat=TRUE)