Robust multivariate location and scatter estimators
This function computes robust estimators for multivariate location and scatter.
covRob(X, type = "auto", maxit = 50, tol = 1e-04, cor = FALSE)
X |
a data matrix with observations in rows. |
type |
a string indicating which estimator to compute. Valid options are "Rocke" for Rocke's S-estimator, "MM" for an MM-estimator with a SHR rho function, or "auto" (default) which selects "Rocke" if the number of variables is greater than or equal to 10, and "MM" otherwise. |
maxit |
Maximum number of iterations, defaults to 50. |
tol |
Tolerance for convergence, defaults to 1e-4. |
cor |
A logical value. If |
This function computes robust estimators for multivariate location and scatter.
The default behaviour (type = "auto") computes a "Rocke" estimator
(as implemented in covRobRocke) if the number
of variables is greater than or equal to 10, and an MM-estimator with a
SHR rho function (as implemented in covRobMM) otherwise.
A list with class “covClassic” with the following components:
mu |
The location estimate |
V |
The scatter matrix estimate, scaled for consistency at the normal distribution |
center |
The location estimate. Same as |
cov |
The scatter matrix estimate, scaled for consistency at the normal distribution. Same as |
cor |
The correlation matrix estimate, if the argument |
dist |
Robust Mahalanobis distances |
Ricardo Maronna, rmaronna@retina.ar
data(bus) X0 <- as.matrix(bus) X1 <- X0[,-9] tmp <- covRob(X1) round(tmp$cov[1:10, 1:10], 3) tmp$mu
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.