order_maxmin: Maximum minimum distance ordering

In GpGp: Fast Gaussian Process Computation Using Vecchia's Approximation

Description

Return the indices of an approximation to the maximum minimum distance ordering. A point in the center is chosen first, and then each successive point is chosen to maximize the minimum distance to previously selected points

Usage

order_maxmin(locs, lonlat = FALSE, space_time = FALSE, st_scale = NULL)

Arguments

locs	A matrix of locations. Each row of locs contains a location, which can be a point in Euclidean space R^d, a point in space-time R^d x T, a longitude and latitude (in degrees) giving a point on the sphere, or a longitude, latitude, and time giving a point in the sphere-time domain.
lonlat	TRUE/FALSE whether locations are longitudes and latitudes.
space_time	TRUE if locations are euclidean space-time locations, FALSE otherwise. If set to TRUE, temporal dimension is ignored.
st_scale	two-vector giving the amount by which the spatial and temporal coordinates are scaled. If NULL, the function uses the locations to automatically select a scaling. If set to FALSE, temporal dimension treated as another spatial dimension (not recommended).

Value

A vector of indices giving the ordering, i.e. the first element of this vector is the index of the first location.

Examples

# planar coordinates
nvec <- c(50,50)
locs <- as.matrix( expand.grid( 1:nvec[1]/nvec[1], 1:nvec[2]/nvec[2] ) )
ord <- order_maxmin(locs)
par(mfrow=c(1,3))
plot( locs[ord[1:100],1], locs[ord[1:100],2], xlim = c(0,1), ylim = c(0,1) )
plot( locs[ord[1:300],1], locs[ord[1:300],2], xlim = c(0,1), ylim = c(0,1) )
plot( locs[ord[1:900],1], locs[ord[1:900],2], xlim = c(0,1), ylim = c(0,1) )

# longitude/latitude coordinates (sphere)
latvals <- seq(-80, 80, length.out = 40 )
lonvals <- seq( 0, 360, length.out = 81 )[1:80]
locs <- as.matrix( expand.grid( lonvals, latvals ) )
ord <- order_maxmin(locs, lonlat = TRUE)
par(mfrow=c(1,3))
plot( locs[ord[1:100],1], locs[ord[1:100],2], xlim = c(0,360), ylim = c(-90,90) )
plot( locs[ord[1:300],1], locs[ord[1:300],2], xlim = c(0,360), ylim = c(-90,90) )
plot( locs[ord[1:900],1], locs[ord[1:900],2], xlim = c(0,360), ylim = c(-90,90) )

GpGp documentation built on June 10, 2021, 1:07 a.m.