sm (version 2.2-6.0)

sm.sphere: Nonparametric density estimation for spherical data.

Description

This function creates a density estimate from data which can be viewed as lying on the surface of a sphere. Directional data form a principal example. The data are displayed in spherical form and a density estimate may be superimposed. The angle of view may be altered. An interactive panel is available to control some features of the estimate and the display. Only modest amounts of data may be used. The limit will depend on the memory available.

Usage

sm.sphere(lat, long, kappa = 20, hidden = FALSE, sphim = FALSE,
          addpoints = FALSE, ...)

Value

a list containing the value of the smoothing parameter and the rotation angles of the displayed plot.

Arguments

lat

a vector giving the latitude component of the data in degrees from the equator.

long

a vector giving the longitude component of the data in degrees east.

kappa

the smoothing parameter used to construct the density estimate. The kernel function is a Fisher distribution and kappa is its scale parameter. Larger values of kappa will produce smaller amounts of smoothing.

hidden

a logical value which, when set to TRUE, will display the points which lie on the rear side of the displayed sphere. This argument will be ignored if sphim is set to TRUE.

sphim

a logical value which controls whether a density estimate is constructed and displayed on the sphere in image form.

addpoints

a logical value which controls whether the data points are added to the plot of the density estimate.

...

arguments for sm.options.

Side Effects

none.

Details

see Section 1.5 of the reference below.

References

Bowman, A.W. and Azzalini, A. (1997). Applied Smoothing Techniques for Data Analysis: the Kernel Approach with S-Plus Illustrations. Oxford University Press, Oxford.

See Also

sm.density

Examples

lat  <- rnorm(50, 10, 15)
long <- c(rnorm(25, 300, 15), rnorm(25, 240, 15))
par(mfrow=c(1,2))
sm.sphere(lat, long)
sm.sphere(lat, long, sphim=TRUE, kappa=15)
par(mfrow=c(1,1))