English

Sub-Gaussian estimators of the mean of a random vector

Statistics Theory 2017-02-03 v1 Machine Learning Statistics Theory

Abstract

We study the problem of estimating the mean of a random vector XX given a sample of NN independent, identically distributed points. We introduce a new estimator that achieves a purely sub-Gaussian performance under the only condition that the second moment of XX exists. The estimator is based on a novel concept of a multivariate median.

Keywords

Cite

@article{arxiv.1702.00482,
  title  = {Sub-Gaussian estimators of the mean of a random vector},
  author = {Gábor Lugosi and Shahar Mendelson},
  journal= {arXiv preprint arXiv:1702.00482},
  year   = {2017}
}

Comments

12 pages