Directional (version 2.4)

Cross validation in von Mises-Fisher discrminant analysis: Cross validation for estimating the classification rate of a discrminant analysis for directional data assuming a von Mises-Fisher distribution

Description

Cross validation for estimating the classification rate of a discrminant analysis for directional data assuming a von Mises-Fisher distribution.

Usage

vmf.da(x, ina, fraction = 0.2, R = 1000, seed = FALSE)

Arguments

x

A matrix with the dataset in Eulcidean coordinates, i.e. unit vectors.

ina

A variable indicating the groupings.

fraction

The fraction of data to be used as test set.

R

The number of repetitions.

seed

If seed is TRUE, the results will always be the same.

Value

A list including: A list including:

Details

A repeated cross validation procedure is performed to estimate the rate of correct classification.

References

Morris, J. E., & Laycock, P. J. (1974). Discriminant analysis of directional data. Biometrika, 61(2): 335-341.

See Also

vmfda.pred, mix.vmf, vmf, dirknn

Examples

x <- rvmf(100, rnorm(4), 15)
ina <- rep(1:2, each = 50)
vmf.da(x, ina, fraction = 0.2, R = 200, seed = FALSE)