Spatial Sign Correlation

TL;DR

This paper introduces a new robust correlation estimator using the spatial sign covariance matrix, analyzing its statistical properties and robustness through theoretical derivations and numerical simulations.

Contribution

It proposes a novel correlation estimator based on SSCM, with detailed asymptotic and robustness analysis, and compares it to existing methods.

Findings

Estimator has strong robustness properties

Asymptotic distribution derived for elliptical distributions

Performs well in finite samples compared to other estimators

Abstract

A new robust correlation estimator based on the spatial sign covariance matrix (SSCM) is proposed. We derive its asymptotic distribution and influence function at elliptical distributions. Finite sample and robustness properties are studied and compared to other robust correlation estimators by means of numerical simulations.