English

Limit theorems of Chatterjee's rank correlation

Statistics Theory 2025-06-05 v4 Machine Learning Probability Statistics Theory

Abstract

Establishing the limiting distribution of Chatterjee's rank correlation for a general, possibly non-independent, pair of random variables has been eagerly awaited by many. This paper shows that (a) Chatterjee's rank correlation is asymptotically normal as long as one variable is not a measurable function of the other, (b) the corresponding asymptotic variance is uniformly bounded by 36, and (c) a consistent variance estimator exists. Similar results also hold for Azadkia-Chatterjee's graph-based correlation coefficient, a multivariate analogue of Chatterjee's original proposal. The proof is given by appealing to H\'ajek representation and Chatterjee's nearest-neighbor CLT.

Keywords

Cite

@article{arxiv.2204.08031,
  title  = {Limit theorems of Chatterjee's rank correlation},
  author = {Zhexiao Lin and Fang Han},
  journal= {arXiv preprint arXiv:2204.08031},
  year   = {2025}
}

Comments

Multiple minor improvements were made in this version, including (1) a proof of the existence of the limiting variance, (2) some numeric studies, and (3) an analysis of the Sobol' indices

R2 v1 2026-06-24T10:50:23.816Z