[2504.08428] Standardization of Weighted Ranking Correlation Coefficients

Statistics > Methodology arXiv:2504.08428 (stat) [Submitted on 11 Apr 2025 (v1), last revised 1 Mar 2026 (this version, v2)] Title:Standardization of Weighted Ranking Correlation Coefficients Authors:Pierangelo Lombardo View a PDF of the paper titled Standardization of Weighted Ranking Correlation Coefficients, by Pierangelo Lombardo View PDF HTML (experimental) Abstract:A fundamental problem in statistics is measuring the correlation between two rankings of a set of items. Kendall's $\tau$ and Spearman's $\rho$ are well established correlation coefficients whose symmetric structure guarantees zero expected value between two rankings randomly chosen with uniform probability. In many modern applications, however, greater importance is assigned to top-ranked items, motivating weighted variants of these coefficients. Such weighting schemes generally break the symmetry of the original formulations, resulting in a non-zero expected value under independence and compromising the interpretation of zero correlation. We propose a general standardization function $g(\cdot)$ that transforms a ranking correlation coefficient $\Gamma$ into a standardized form $g(\Gamma)$ with zero expected value under randomness. The transformation preserves the domain $[-1,1]$, satisfies the boundary conditions, is continuous and increasing, and reduces to the identity for coefficients that already satisfy the zero-expected-value property. The construction of $g(x)$ depends on three distributional para...