Description Usage Arguments Details Value Author(s) See Also Examples
A function to estimate home ranges with kernel density estimation.
1 2 3 4 5 |
xy |
|
levels |
numeric vector with the isopleth levels. |
h |
character ("href" or "hlscv") specifying the method to estimate the bandwidth or numeric value specifying the bandwidth. |
xrange |
numeric vector specifying min and max x for the output grid. |
yrange |
numeric vector specifying min and max y for the output grid. |
increaseExtent |
numeric value by which the x and y
range are extended, see also |
trast |
a |
buffer |
numeric value to buffer the bounding box of points in map units from which the extent of the output grid will be retrieved. |
res |
numeric value specifying the resolution for the output grid. |
gridsize |
a vector of length 2, specifying the number of rows and the number of columns for the ouput grid. |
ud |
logical value, indicating whether or not a utilization distribution should be calculated. |
cud |
logical value, indicating whether or not a cumulative utilization distribution should be calculated. |
rescale |
character value specifying if the data
should be rescaled before calculating bandwidth. Possible
values are: |
lscvSearch |
numeric vector of length 2, specifying lower and uper bound for candidate bandwidth (as portion of reference bandwidth) for estimating bandwidth with least squre cross validation. |
lscvWhichMin |
character value, specifying how
candidate bandwidths are chosen with least squre cross
validation. Possible values are: |
The size and resolution of the resulting utilization
distribution (UD) grid is influenced by traster,
xrange, yrange, increaseExtent, buffer, res, gridsize
.
The size of the grid can be set either through a template
raster (traster
), xrange
and yrange
or increaseExtent
. traster
takes precedence
over xrange
and yrange
, buffer
and
grid
. If none of the previous arguments are
provided, xrange
and yrange
are taken from
the data.
The resolution of the resulting UD grid can be set
through either res
or gridsize
. res
takes precedence over gridsize
. If none of the
previous arguments is provided the grid is set by default
to a 100 by 100 grid. The bandwidth can be provided by
the user or estimated through the reference bandwidth
(this method is often refered to as the ad hoc method),
plug in the euqtion method or the least square cross
validation method. Reference bandwidth estimation is
implemented as suggested by Silverman 1986. Plugin the
equation method is wrapped from KernSmooth::dpki
and a simple binned version Silverman's suggestion for
least square cross validation is implemented. Kernels
densities are estimated with KernSmooth::bkde2d
.
This is a binned approximation of 2D kernel density
estimates (see ?KernSmooth::bkde2d
for more
details.
object of class RhrHREstimator
Johannes Signer
KernSmooth::bkde2d
, KernSmooth::dpik
,
rhr::rhrHref
, rhr::rhrHlscv
,
rhr::rhrHpi
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | data(datSH)
## Not run:
# Kernel with href bandwidth estimation
k1 <- rhrKDE(datSH[, 2:3], h="href", res=100)
plot(k1)
# what is the actually estimated bandwidth?
k1$parameters$h
# Kernel with href bandwidth estimation
k2 <- rhrKDE(datSH[, 2:3], h="lscv", res=100)
plot(k2)
# what is the actually estimated bandwidth?
k2$parameters$h
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.