On the asymptotic distribution of the symmetrized Chatterjee's correlation coefficient

TL;DR

This paper derives the asymptotic distribution of the symmetric Chatterjee's correlation coefficient and proposes a finite sample independence test, advancing statistical tools for dependence measurement.

Contribution

It provides the first derivation of the asymptotic distribution for the symmetric version of Chatterjee's correlation and introduces a practical independence testing method.

Findings

Asymptotic distribution derived for symmetric Chatterjee's correlation

Finite sample test for independence proposed

Enhances statistical dependence measurement tools

Abstract

Chatterjee (2021) introduced an asymmetric correlation measure that has attracted much attention over the past year. In this paper, we derive the asymptotic distribution of the symmetric version of Chatterjee's correlation, and suggest a finite sample test for independence.