Minimum Volume Ellipsoid covariance estimator
This function uses a fast algorithm to compute the Minimum Volume Ellipsoid (MVE) for multivariate location and scatter.
fastmve(x, nsamp = 500)
x |
data matrix (n x p) with cases stored in rows. |
nsamp |
number of random starts for the iterative algorithm, these are constructed using subsamples of the data. |
This function computes the Minimum Volume
Ellipsoid (MVE) for multivariate location and scatter, using a
fast algorithm related to the fast algorithm for S-regression
estimators (see lmrob).
A list with the following components:
center |
a vector with the robust multivariate location estimate |
cov |
a matrix with the robust covariance / scatter matrix estimate |
scale |
A scalar that equals the median of the mahalanobis distances of the
data to the |
best |
Indices of the observations that correspond to the MVE estimator |
nsamp |
Number of random starts used for the iterative algorithm |
nsing |
Number of random subsamples (among the |
Matias Salibian-Barrera, matias@stat.ubc.ca
data(bus) X0 <- as.matrix(bus) X1 <- X0[,-9] tmp <- fastmve(X1) round(tmp$cov[1:10, 1:10], 3) tmp$center
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.