powered by
This function computes robust multivariate location and scatter estimators using both random and deterministic starting points.
initPP(X, muldirand = 20, muldifix = 10, dirmin = 1000)
A list with the following components:
A zero/one vector with ones in the positions of the suspected outliers
Robust squared Mahalanobis distances
Robust mean estimate
Robust covariance matrix estimate
Outlyingness of data points
a data matrix with observations in rows.
used to determine the number of random directions (candidates), which is max(p*muldirand, dirmin), where p is the number of columns in X.
max(p*muldirand, dirmin)
p
X
used to determine the number of random directions (candidates), which is min(n, 2*muldifix*p).
min(n, 2*muldifix*p)
minimum number of random directions
Ricardo Maronna, rmaronna@retina.ar, based on original code by D. Pen~a and J. Prieto
This function computes robust multivariate location and scatter using both Pen~a-Prieto and random candidates.
http://www.wiley.com/go/maronna/robust
data(bus) X0 <- as.matrix(bus) X1 <- X0[,-9] tmp <- initPP(X1) round(tmp$cov[1:10, 1:10], 3) tmp$center
Run the code above in your browser using DataLab