Learn R Programming

amt (version 0.2.2.0)

hr_kde_ref_scaled: Select a bandwidth for Kernel Density Estimation

Description

Use two dimensional reference bandwidth to select a bandwidth for kernel density estimation. Find the smallest value for bandwidth (h) that results in n polygons (usually n=1) contiguous polygons at a given level.

Usage

hr_kde_ref_scaled(
  x,
  range = hr_kde_ref(x)[1] * c(0.01, 1),
  trast = make_trast(x),
  num.of.parts = 1,
  levels = 0.95,
  tol = 0.1,
  max.it = 500L
)

Value

list with the calculated bandwidth, exit status and the number of iteration.

Arguments

x

A track_xy*.

range

Numeric vector, indicating the lower and upper bound of the search range. If range is to large with regard to trast, the algorithm will fail.

trast

A template RasterLayer.

num.of.parts

Numeric numeric scalar, indicating the number of contiguous polygons desired. This will usually be one.

levels

The home range level.

tol

Numeric scalar, indicating which difference of to stop.

max.it

Numeric scalar, indicating the maximum number of acceptable iterations.

Details

This implementation uses a bisection algorithm to the find the smallest value value for the kernel bandwidth within range that produces an home-range isopleth at level consisting of n polygons. Note, no difference is is made between the two dimensions.

References

Kie, John G. "A rule-based ad hoc method for selecting a bandwidth in kernel home-range analyses." Animal Biotelemetry 1.1 (2013): 1-12.