rrcov (version 0.4-08)

rrcov.control: Control object for the estimation parameters

Description

Useful for passing the estimation options as parameters to the estimation functions

Usage

rrcov.control(alpha=1/2, nsamp=500, seed=NULL, tolSolve=10e-14,
    trace=FALSE, use.correction=TRUE, adjust=FALSE,
    r = 0.45, arp = 0.05, eps=1e-3, maxiter=120)

Arguments

alpha

This parameter controls the size of the subsets over which the determinant is minimized, i.e. alpha*n observations are used for computing the determinant. Allowed values are between 0.5 and 1 and the default is 0.5.

nsamp

number of subsets used for initial estimates or "best" or "exact". Default is nsamp = 500. If nsamp="best" exhaustive enumeration is done, as far as the number of trials do not exceed 5

seed

starting value for random generator. Default is seed = NULL

tolSolve

numeric tolerance to be used for inversion (solve) of the covariance matrix in mahalanobis.

trace

whether to print intermediate results. Default is trace = FALSE

use.correction

whether to use finite sample correction factors. Default is use.correction=TRUE

adjust

whether to perform intercept adjustment at each step. This could be quite time consuming, therefore the default is adjust = FALSE

r

M-estimates: breakdown point, i.e. the fraction of contaminated data. The default is 0.45

arp

M-estimates: asypmthotic rejection point, i.e. the fraction of points receiving zero weights. The default is 0.001

eps

M-estimates: the relaive precision of the solution. The default is 1e-3

maxiter

M-estimates: maximum number of iterations for the computation of the M-estimates. The default is 120

Value

• A list with components, as the parameters passed by the invocation

Details

For details about the estimation options see the corresponding estimation functions.

Examples

data(Animals, package = "MASS")
brain <- Animals[c(1:24, 26:25, 27:28),]
data(hbk)
hbk.x <- data.matrix(hbk[, 1:3])

ctrl <- rrcov.control(alpha=0.75, trace=TRUE)
covMcd(hbk.x,      control = ctrl)
covMcd(log(brain), control = ctrl)