sKDE: Computes an estimation of the density using Kernel Density...

In ripleyCorr/kdeborder: Kernel density estimation correcting for border bias

Description

Computes an estimation of the density using Kernel Density estimator, correcting for fontier effects.

Usage

sKDE(U, polygon, optimal = TRUE, h = 0.1, parallel = FALSE,
  n_clusters = 4)

Arguments

U	data points.
polygon	polygon on which points lie.
optimal	if TRUE, uses Hpi() to select the optimal bandwidth.
h	only if optimal=FALSE, scalar bandwidth.
parallel	if TRUE, computes the weights using clusters.
n_clusters	only if n_clusters=TRUE, defines the number of clusters.

Value

Returns a list whose elements are: > X: x coordinates at wich estimate is evaluated, > Y: y coordinates at wich estimate is evaluated, > Z: density estimates, > ZNA: density estimates with NA values for points outside the polygon, > H: bandwidth matrix, > W: vector of weights.

References

Charpentier, A. & Gallic, E. (2015). Kernel density estimation based on Ripley’s correction. GeoInformatica, 1-22. (Springer)

See Also

sWeights which this function wraps

Examples

data(acci)
# Estimation with correction
smoothed_fin <- sKDE(U = acci$finistere$points, polygon = acci$finistere$polygon,
optimal=TRUE, parallel = FALSE)

ripleyCorr/kdeborder documentation built on May 26, 2019, 1:33 a.m.