R: Compute a Three Dimension Kernel Density Estimate
kde3d
R Documentation
Compute a Three Dimension Kernel Density Estimate
Description
Evaluates a three dimensional kernel density estimate using a Gaussian
kernel with diagonal covariance matrix on a regular grid.
Usage
kde3d(x, y, z, h, n = 20, lims = c(range(x), range(y), range(z)))
Arguments
x,y,z
x, y, and z coordinates of the data.
h
vector of three bandwidths for the density estimate;
recycled if length is less than three; default is based on the
normal reference bandwidth (see bandwidth.nrd).
n
numbers of grid points to use for each dimension; recycled if
length is less than three.
lims
lower and upper limits on the region for which the density
estimate is to be computed, provides as a vector of length 6,
corresponding to low and high values of x, y, and
z; recycled if only two values are supplied.
Value
A list of four components, x, y, z, and
d. x, y, and z are the coordinates of the
grid points at which the density estimate has been evaluated, and
d is a three dimensional array of the estimated density values.
References
Based on the function kde2d in package
MASS.
See Also
kde2d.
Examples
with(quakes, {
d <- kde3d(long, lat, -depth, n = 40)
contour3d(d$d, exp(-12), d$x/22, d$y/28, d$z/640,
color = "green", color2 = "gray", scale=FALSE,
engine = "standard")
})