Extracts the generalized kurtosis values of the components obtained via an ICS transformation.
gen_kurtosis(object, ...)# S3 method for ICS
gen_kurtosis(object, select = NULL, scale = FALSE, index = NULL, ...)
A numeric vector containing the generalized kurtosis values of the requested components.
an object inheriting from class "ICS" containing
results from an ICS transformation.
additional arguments to be passed down.
an integer, character, or logical vector specifying for which
components to extract the generalized kurtosis values, or NULL to
extract the generalized kurtosis values of all components.
a logical indicating whether to scale the generalized kurtosis
values to have product 1 (default to FALSE). See ‘Details’
for more information.
an integer vector specifying for which components to extract
the generalized kurtosis values, or NULL to extract the generalized
kurtosis values of all components. Note that index is deprecated
and may be removed in the future, use select instead.
Andreas Alfons and Aurore Archimbaud
The argument scale is useful when ICS is performed with shape
matrices rather than true scatter matrices. Let \(S_{1}\) and
\(S_{2}\) denote the scatter or shape matrices used in ICS.
If both \(S_{1}\) and \(S_{2}\) are true scatter matrices, their order in principal does not matter. Changing their order will just reverse the order of the components and invert the corresponding generalized kurtosis values.
The same does not hold when at least one of them is a shape matrix rather than a true scatter matrix. In that case, changing their order will also reverse the order of the components, but the ratio of the generalized kurtosis values is no longer 1 but only a constant. This is due to the fact that when shape matrices are used, the generalized kurtosis values are only relative ones. It is then useful to scale the generalized kurtosis values such that their product is 1.
data("iris")
X <- iris[,-5]
out <- ICS(X)
gen_kurtosis(out)
gen_kurtosis(out, scale = TRUE)
gen_kurtosis(out, select = c(1,4))
Run the code above in your browser using DataLab