English

Affine equivariant rank-weighted L-estimation of multivariate location

Statistics Theory 2015-04-02 v2 Statistics Theory

Abstract

In the multivariate one-sample location model, we propose a class of flexible robust, affine-equivariant L-estimators of location, for distributions invoking affine-invariance of Mahalanobis distances of individual observations. An involved iteration process for their computation is numerically illustrated.

Keywords

Cite

@article{arxiv.1503.05392,
  title  = {Affine equivariant rank-weighted L-estimation of multivariate location},
  author = {Pranab Kumar Sen and Jana Jureckova and Jan Picek},
  journal= {arXiv preprint arXiv:1503.05392},
  year   = {2015}
}

Comments

16 pages, 4 figures, 6 tables