The Spherical Principal Components procedure was proposed by Locantore et al., (1999) as a functional data analysis method. The idea is to perform classical PCA on the the data, \ projected onto a unit sphere. The estimates of the eigenvectors are consistent and the procedure is extremly fast. The simulations of Maronna (2005) show that this method has very good performance.
Objects can be created by calls of the form new("PcaLocantore", ...)
but the
usual way of creating PcaLocantore
objects is a call to the function
PcaLocantore
which serves as a constructor.
delta
:Accuracy parameter
quan
:Object of class "numeric"
The quantile h used throughout the algorithm
call
, center
, scale
, rank
, loadings
,
eigenvalues
, scores
, k
,
sd
, od
, cutoff.sd
, cutoff.od
,
flag
, n.obs
, eig0
, totvar0
:from the "Pca"
class.
Class "PcaRobust"
, directly.
Class "Pca"
, by class "PcaRobust", distance 2.
signature(obj = "PcaLocantore")
: ...
Valentin Todorov valentin.todorov@chello.at
Todorov V & Filzmoser P (2009), An Object Oriented Framework for Robust Multivariate Analysis. Journal of Statistical Software, 32(3), 1--47. tools:::Rd_expr_doi("10.18637/jss.v032.i03").
PcaRobust-class
, Pca-class
, PcaClassic
, PcaClassic-class