Vishal <vishalps <at> gmail.com> writes: > > Jari, thanks for the quick answer. > > > sqrt(eigen(cov.trob(mydataforellipse)$cov)$values) > > what will this return? > > For my data, I get: > > > sqrt(eigen(cov.trob(r)$cov)$values) > [1] 1.857733e-05 4.953181e-06 > > Is this Left hand value the major or the semi major length? > > I also try to plot a circuit keeping this as the radius/diameter > I don't get a circle that intersects the major axis of the ellipse. > Vishal,
Then you probably do not have an ellipse directly defined by the robust covariance matrix, but it is scaled up by some statistic. Check the code of the function (or in a happy case, its documentation) to see what is the scaling factor. I checked with drawing an ellipse directly with the cov.trob data and both of the circles fit nicely. Cheers, Jari Oksanen ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
