On the convergence of Maronna's MM-estimators of scatter

preprint2014arXivOpen access

Abstract

In this paper, {we propose an alternative proof for the uniqueness} of Maronna's MM-estimator of scatter (Maronna, 1976) for NN vector observations y1,...,yN∈Rm\mathbf y_1,...,\mathbf y_N\in\mathbb R^m under a mild constraint of linear independence of any subset of mm of these vectors. This entails in particular almost sure uniqueness for random vectors yi\mathbf y_i with a density as long as N>mN>m. {This approach allows to establish further relations that demonstrate that a properly normalized Tyler's MM-estimator of scatter (Tyler, 1987) can be considered as a limit of Maronna's MM-estimator. More precisely, the contribution is to show that each MM-estimator converges towards a particular Tyler's MM-estimator.} These results find important implications in recent works on the large dimensional (random matrix) regime of robust MM-estimation.

Explore this paper’s authors, topics and related work.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Reviews 0

Write a reviewWrite

No reviews yet.

Discussion 0

Add a commentWrite

No comments yet.