Directional (version 4.0)

von Mises kernel density estimation: Kernel density estimation of circular data with a von Mises kernel

Description

Kernel density estimation of circular data with a von Mises kernel.

Usage

vm.kde(u, h, thumb = "none", rads = TRUE)

Arguments

u

A numeric vector containing the data.

h

The bandwidth.

thumb

It can be either "none", so the bandwidth the user has set will be used, "tay" for the method of Taylor (2008) or "rot" for the method of Garcia-Portugues (2013).

rads

If the data are in radians, this should be TRUE and FALSE otherwise.

Value

A list including:

h

The bandwidth. If the user chose one of "tay" or "rot" the estimated bandwidth will be returned.

f

The kernel density estimate at the observed points.

Details

The user has the option to use a bandwidth he/she has found in some way (cross-validation) or estimate it as Taylor (2008) or Garcia-Portugues (2013).

References

Taylor, C. C. (2008). Automatic bandwidth selection for circular density estimation. Computational Statistics & Data Analysis, 52(7): 3493-3500.

Garcia Portugues, E. (2013). Exact risk improvement of bandwidth selectors for kernel density estimation with directional data. Electronic Journal of Statistics, 7, 1655-1685.

See Also

vmkde.tune, vmfkde.tune, vmf.kde

Examples

# NOT RUN {
x <- rvonmises(100, 2.4, 10, rads = TRUE)
hist(x, freq = FALSE)
f1 <- vm.kde(x, h = 0.1, thumb = "rot", rads = TRUE)$f
f2 <- vm.kde(x, h = 0.1, thumb = "tay", rads = TRUE)$f
h <- vmkde.tune(x)[1]
f3 <- vm.kde(x, h = h, thumb = "none", rads = TRUE)$f
points(x, f1, col = 1)
points(x, f2, col = 2)
points(x, f3, col = 3)
# }