Circular distance correlation between two circular variables {Directional}

Circular distance correlation between two circular variables

Description

Circular distance correlation between two circular variables.

Usage

circ.dcor(theta, phi, rads = FALSE)

Arguments

theta

The first cirular variable.

phi

The other cirular variable.

rads

If the data are expressed in rads, then this should be TRUE. If the data are in degrees, then this is FALSE.

Details

The angular data are transformed to their Euclidean coordinates and then the distance correlation is computed.

Value

A list including:

dcov

The distance covariance.

dvarX

The distance variance of x.

dvarY

The distance variance of Y.

dcor

The distance correlation.

Author(s)

Michail Tsagris.

R implementation and documentation: Michail Tsagris [e-mail].

References

G.J. Szekely, M.L. Rizzo and N. K. Bakirov (2007). Measuring and Testing Independence by Correlation of Distances. Annals of Statistics, 35(6):2769-2794.

See Also

circlin.cor, circ.cor2, spher.dcor

Examples

y <- runif(50, 0, 2 * pi)
x <- runif(50, 0, 2 * pi)
circ.dcor(x, y, rads = TRUE)