truncPC: Classical Principal Components by truncated SVD.

Description

Similar usage to robustbase::classPC except for the new argument ncomb which is the desired number of components. Only this many PC's are computed in order to save computation time. Makes use of propack.svd of package svd.

Usage

truncPC(X, ncomp = NULL, scale = FALSE, center = TRUE, 
                    signflip = TRUE, via.svd = NULL, scores = FALSE)

Value

A list with components:

rank: the (numerical) matrix rank of X, i.e. an integer number between 0 and min(dim(x)).
eigenvalues: the k eigenvalues, proportional to the variances, where k is the rank above.
loadings: the loadings, a \(d \times k\) matrix.
scores: if the scores argument was TRUE, the \(n \times k\) matrix of scores.
center: a vector of means, unless the center argument was FALSE.
scale: a vector of column scales, unless the scale argument was false.

Arguments

X: a numeric matrix.
ncomp: the desired number of components (if not specified, all components are computed).
scale: logical, or numeric vector for scaling the columns.
center: logical or numeric vector for centering the matrix.
signflip: logical indicating if the signs of the loadings should be flipped such that the absolutely largest value is always positive.
via.svd: dummy argument for compatibility with classPC calls, will be ignored.
scores: logical indicating whether or not scores should be returned.

Author

P.J. Rousseeuw

Examples

Run this code

library(MASS) 
set.seed(12345) 
n <- 100; d <- 10
A <- diag(d) * 0.1 + 0.9
x <- mvrnorm(n, rep(0,d), A)
truncPCA.out <- truncPC(x, ncomp = 2, scores = TRUE)
plot(truncPCA.out$scores)