Robpca-class {rrcov} | R Documentation |
The ROBPCA algorithm was proposed by Hubert et al (2005) and stays for 'ROBust method for Principal Components Analysis'. It is resistant to outliers in the data. The robust loadings are computed using projection-pursuit techniques and the MCD method. Therefore ROBPCA can be applied to both low and high-dimensional data sets. In low dimensions, the MCD method is applied.
Objects can be created by calls of the form new("Robpca", ...)
but the
usual way of creating Robpca
objects is a call to the function
Robpca
which serves as a constructor.
alpha
:"numeric"
the fraction of outliers
the algorithm should resist - this is the argument alphaquan
:"numeric"
The quantile h used throughout the algorithm call
:"language"
center
:"vector"
the center of the data loadings
:"matrix"
the matrix
of variable loadings (i.e., a matrix whose columns contain the eigenvectors) eigenvalues
:"vector"
the eigenvalues scores
:"matrix"
the scores - the value
of the rotated data (the centred (and scaled if requested) data multiplied
by the rotation
matrix) is returned. Hence, cov(scores)
is the diagonal matrix diag(eigenvalues)
k
:"numeric"
number of (choosen) principal components sd
:"Uvector"
Score distances within the robust PCA subspace od
:"Uvector"
Orthogonal distances to the robust PCA subspace cutoff.sd
:"numeric"
Cutoff value for the score distancescutoff.od
:"numeric"
Cutoff values for the orthogonal distances flag
:"Uvector"
The observations whose score distance is larger
than cutoff.sd or whose orthogonal distance is larger than cutoff.od can be considered
as outliers and receive a flag equal to zero.
The regular observations receive a flag 1 n.obs
:"numeric"
the number of observations
Class "PcaRobust"
, directly.
Class "Pca"
, by class "PcaRobust", distance 2.
signature(obj = "Robpca")
: Returns the quantile
used throughout the algorithm Valentin Todorov valentin.todorov@chello.at
PcaRobust-class
, Pca-class
, PcaClassic
, PcaClassic-class
showClass("Robpca")