SMPCA: Robust principal components

In msalibian/RobStatTM: Robust Statistics: Theory and Methods

Robust principal components

Description

This function computes robust principal components based on the minimization of the "residual" M-scale.

Usage

pcaRobS(X, ncomp, desprop = 0.9, deltasca = 0.5, maxit = 100)

Arguments

X	a data matrix with observations in rows.
ncomp	desired (maximum) number of components
desprop	desired (minimum) proportion of explained variability (default = 0.9)
deltasca	"delta" parameter of the scale M-estimator (default=0.5)
maxit	maximum number of iterations (default= 100)

Value

A list with the following components:

q	The actual number of principal components
propex	The actual proportion of unexplained variability
eigvec	Eigenvectors, in a p x q matrix
fit	an n x p matrix with the rank-q approximation to X
repre	An n x q matrix with representation of data in R^q (scores)
propSPC	A vector of length p with the cumulative explained variance from initial SPC

Author(s)

Ricardo Maronna, [e-mail], based on original code by D. Pen~a and J. Prieto

References

http://www.wiley.com/go/maronna/robust

Examples

data(bus)
X0 <- as.matrix(bus)
X1 <- X0[,-9]
ss <- apply(X1, 2, mad)
mu <- apply(X1, 2, median)
X <- scale(X1, center=mu, scale=ss)
q <- 3  #compute three components
rr <- pcaRobS(X, q, 0.99)
round(rr$eigvec, 3)

msalibian/RobStatTM documentation built on June 13, 2025, 12:45 p.m.