English

A note on breaking ties among sample medians

Methodology 2019-09-04 v2

Abstract

Given samples x1,,xnx_1,\cdots,x_n, it is well known that any sample median value (not necessarily unique) minimizes the absolute loss i=1nqxi\sum_{i=1}^n |q-x_i|. Interestingly, we show that the minimizer of the loss i=1nqxi1+ϵ\sum_{i=1}^n|q-x_i|^{1+\epsilon} exhibits a singular perturbation behaviour that provides a unique definition for the sample median as ϵ0\epsilon \rightarrow 0. This definition is the unique point among all candidate median values that balances the logarithmiclogarithmic moment of the empirical distribution. The result generalizes directly to breaking ties among sample quantiles when the quantile regression loss is modified in the same way.

Cite

@article{arxiv.1807.03462,
  title  = {A note on breaking ties among sample medians},
  author = {Peter M. Aronow and Donald K. K. Lee},
  journal= {arXiv preprint arXiv:1807.03462},
  year   = {2019}
}

Comments

5 pages, no tables or figures

R2 v1 2026-06-23T02:55:50.043Z