tolellipse {rrcov}R Documentation

Tolerance Ellipse Plot


Plots the 0.975 tolerance ellipse of the bivariate data set x. The ellipse is defined by those data points whose distance is equal to the squareroot of the 0.975 chisquare quantile with 2 degrees of freedom.


tolellipse(x, mcd, cutoff, id.n, classic = FALSE, tol.inv=1e-7)


x a two dimensional matrix or data frame.
mcd An object of type mcd - its attributes center and cov will be used. If missing the MCD will be computed.
cutoff Distance needed to flag data points outside the ellipse
id.n Number of observations to identify by a label. If not supplied, the number of observations with distance larger than cutoff is used.
classic Whether to plot the classical distances too. Default is classic=FALSE.
tol.inv Tolerance to be used for computing the inverse see 'solve'. Default is tol.inv = 10e-7

See Also



mcd <- covMcd(hbk.x)       # compute mcd in advance
tolellipse(hbk.x[,1:2])    # must be a 2-dimensional data set: take the first two columns

[Package rrcov version 0.2-5 Index]