The Null Variance of Spearman's Rho under the Bivariate Student t: Monotonicity, a Weight-Five Core at Odd ν, and a Pitman Crossover with Kendall's Tau

This repository presents the monograph, high-precision reproduction script, machine-readable JSON verification dataset, and the updated 28-suite K-PROTOCOL Master Catalog & Official Repository Downloader for "The Null Variance of Spearman's Rho under the Bivariate Student t: Monotonicity, a Weight-Five Core at Odd nu, and a Pitman Crossover with Kendall's Tau" (K-PROTOCOL Suite 28; Permanent Concept DOI: 10.5281/zenodo.24414005). Under the uncorrelated (r = 0) bivariate Student t_nu distribution, the two coordinates remain dependent through the common scale, so the asymptotic null variance sigma_S^2(nu) = lim n * Var(rho_S_hat) of Spearman's rho exceeds 1. This paper derives its exact structure across all nu > 0 and compares the resulting Pitman efficacy with Pearson's correlation and Kendall's tau: - Proposition 1 & Theorem 1 (Exact Formula and Strict Monotonicity): Reflection symmetry reduces the null variance to sigma_S^2(nu) = 144 * E[(F_nu(X) - 1/2)^2 * (F_nu(Y) - 1/2)^2]. Via a quantile log-derivative ordering for Gamma(a, 1) (Lemma 1), Hoeffding's covariance identity, and Fatou's lemma, the copula of (|X|, |Y|) decreases pointwise in nu and sigma_S^2(nu) decreases strictly with strictly negative derivative from 9/5 (nu -> 0) to 1 (nu -> infinity).- Theorem 2 (Even Degrees of Freedom): For every integer m >= 1, sigma_S^2(2m) lies in Q + Q*pi (e.g., sigma_S^2(2) = 9*pi - 27, with exact closed forms verified through nu = 8).- Theorems 3 & 4 & Proposition 2 (Odd Degrees of Freedom and Invariant Weight-Five Core): For the Cauchy law (nu = 1), sigma_S^2(1) reduces to a one-dimensional inverse tangent integral and evaluates analytically via alternating Euler sums to S_1 = -23/5 - 48*ln^2(2)/pi^2 + 48*ln^4(2)/pi^4 + 252*zeta(3)*ln(2)/pi^4 + 1152*Li_4(1/2)/pi^4. By a 2D divergence identity (Lemma 2) and boundary parity cancellation (Appendix A, Lemmas 4-5), for every odd nu = 2m - 1 the weight-five core S_1 is completely invariant and nu moves only the four constants pi^{-2}, pi^{-4}, zeta(3)*pi^{-4}, and ln(2)*pi^{-4} (identified to 85-114 digits for nu = 3, 5, 7).- Section 6 & Section 7 (Pitman Efficacy, Kendall's Tau Monotonicity, and Crossover): For nu = 2, e_S(2) = (pi^2 - 8)^2 / (4*pi^2*(pi - 3)); against Pearson's test, Spearman's test is more efficient for 5 <= nu <= 17. Expressing Kendall's null variance as sigma_tau^2(nu) = (16/pi^2) E[arcsin^2(V_tilde)] (Proposition 3, matching Dengler 2010, whose odd-nu closed forms carry an invariant weight-two core 16*ln^2(2)/pi^2) yields strict monotonicity on (0, infinity) from 4/3 to 4/9 as a corollary of Suite 27 (Corollary 1). By Lemma 3 and Theorem 5, ARE(tau, S; nu) tends to 27/20 as nu -> 0 and satisfies 1 + (1/4 - 3/pi^2)/nu + O(nu^{-2}) as nu -> infinity. Because pi^2 < 12, the two tests cross at a finite nu_* approx 4.42 (Kendall is more efficient for nu <= 4, Spearman for nu >= 5, with a minimum near 0.99730 at nu approx 9.5). REPOSITORY CONTENTS:- K-PROTOCOL_Spearman_Null_Variance.pdf : Complete 11-page paper with proofs, Appendix A (proof of Theorem 4), exact closed-form tables, Pitman efficacy and ARE comparisons against Pearson and Kendall.- verify_spearman_null_variance.py : Self-contained Python script (standard library, numpy, scipy) verifying closed forms for nu = 1..8 against 60-digit double-exponential quadrature (diff <= 2e-28), Cauchy 1D formula (diff 4.5e-43), every step of Theorem 3 and Appendix A, Gamma quantile ordering and copula monotonicity, seeded Monte Carlo checks, 60-digit Kendall quadrature vs. Dengler (2010) (diff <= 1.4e-51), Richardson 1/nu expansion coefficients (nu = 400, 800), and the crossover bracket.- spearman_null_variance_reproduction.json : Machine-readable JSON output reproducing all numerical checks, tables, Monte Carlo runs, and crossover diagnostics (all_pass: true).- K-PROTOCOL Master Series Catalog and 28-Suite DOI Index.pdf : Updated 28-suite architectural blueprint and permanent Concept DOI directory (Suites 01-28), including Division IV (Mathematical Statistics & Information Geometry).- k_protocol_official_zenodo_repository_downloader_28_suites.py : Official Python repository downloader supporting selective retrieval of Foundational Roots (01-03), All 28 Suites (01-28), or Mathematical Statistics Suites (24-28). Division IV: Mathematical Statistics & Information Geometry

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-21
DOI
https://doi.org/10.5281/zenodo.24414005
Citations
1
Primary Topic
Probability and Statistical Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

The Null Variance of Spearman's Rho under the Bivariate Student t: Monotonicity, a Weight-Five Core at Odd ν, and a Pitman Crossover with Kendall's Tau

A Citizen of the Republic of Korea
1 citations
Zenodo (CERN European Organization for Nuclear Research)
Probability and Statistical Research
preprint

The Null Variance of Spearman's Rho under the Bivariate Student t: Monotonicity, a Weight-Five Core at Odd ν, and a Pitman Crossover with Kendall's Tau

A Citizen of the Republic of Korea
preprint en
1 citations

Abstract

This repository presents the monograph, high-precision reproduction script, machine-readable JSON verification dataset, and the updated 28-suite K-PROTOCOL Master Catalog & Official Repository Downloader for "The Null Variance of Spearman's Rho under the Bivariate Student t: Monotonicity, a Weight-Five Core at Odd nu, and a Pitman Crossover with Kendall's Tau" (K-PROTOCOL Suite 28; Permanent Concept DOI: 10.5281/zenodo.24414005). Under the uncorrelated (r = 0) bivariate Student t_nu distribution, the two coordinates remain dependent through the common scale, so the asymptotic null variance sigma_S^2(nu) = lim n * Var(rho_S_hat) of Spearman's rho exceeds 1. This paper derives its exact structure across all nu > 0 and compares the resulting Pitman efficacy with Pearson's correlation and Kendall's tau: - Proposition 1 & Theorem 1 (Exact Formula and Strict Monotonicity): Reflection symmetry reduces the null variance to sigma_S^2(nu) = 144 * E[(F_nu(X) - 1/2)^2 * (F_nu(Y) - 1/2)^2]. Via a quantile log-derivative ordering for Gamma(a, 1) (Lemma 1), Hoeffding's covariance identity, and Fatou's lemma, the copula of (|X|, |Y|) decreases pointwise in nu and sigma_S^2(nu) decreases strictly with strictly negative derivative from 9/5 (nu -> 0) to 1 (nu -> infinity).- Theorem 2 (Even Degrees of Freedom): For every integer m >= 1, sigma_S^2(2m) lies in Q + Q*pi (e.g., sigma_S^2(2) = 9*pi - 27, with exact closed forms verified through nu = 8).- Theorems 3 & 4 & Proposition 2 (Odd Degrees of Freedom and Invariant Weight-Five Core): For the Cauchy law (nu = 1), sigma_S^2(1) reduces to a one-dimensional inverse tangent integral and evaluates analytically via alternating Euler sums to S_1 = -23/5 - 48*ln^2(2)/pi^2 + 48*ln^4(2)/pi^4 + 252*zeta(3)*ln(2)/pi^4 + 1152*Li_4(1/2)/pi^4. By a 2D divergence identity (Lemma 2) and boundary parity cancellation (Appendix A, Lemmas 4-5), for every odd nu = 2m - 1 the weight-five core S_1 is completely invariant and nu moves only the four constants pi^{-2}, pi^{-4}, zeta(3)*pi^{-4}, and ln(2)*pi^{-4} (identified to 85-114 digits for nu = 3, 5, 7).- Section 6 & Section 7 (Pitman Efficacy, Kendall's Tau Monotonicity, and Crossover): For nu = 2, e_S(2) = (pi^2 - 8)^2 / (4*pi^2*(pi - 3)); against Pearson's test, Spearman's test is more efficient for 5 <= nu <= 17. Expressing Kendall's null variance as sigma_tau^2(nu) = (16/pi^2) E[arcsin^2(V_tilde)] (Proposition 3, matching Dengler 2010, whose odd-nu closed forms carry an invariant weight-two core 16*ln^2(2)/pi^2) yields strict monotonicity on (0, infinity) from 4/3 to 4/9 as a corollary of Suite 27 (Corollary 1). By Lemma 3 and Theorem 5, ARE(tau, S; nu) tends to 27/20 as nu -> 0 and satisfies 1 + (1/4 - 3/pi^2)/nu + O(nu^{-2}) as nu -> infinity. Because pi^2 < 12, the two tests cross at a finite nu_* approx 4.42 (Kendall is more efficient for nu <= 4, Spearman for nu >= 5, with a minimum near 0.99730 at nu approx 9.5). REPOSITORY CONTENTS:- K-PROTOCOL_Spearman_Null_Variance.pdf : Complete 11-page paper with proofs, Appendix A (proof of Theorem 4), exact closed-form tables, Pitman efficacy and ARE comparisons against Pearson and Kendall.- verify_spearman_null_variance.py : Self-contained Python script (standard library, numpy, scipy) verifying closed forms for nu = 1..8 against 60-digit double-exponential quadrature (diff <= 2e-28), Cauchy 1D formula (diff 4.5e-43), every step of Theorem 3 and Appendix A, Gamma quantile ordering and copula monotonicity, seeded Monte Carlo checks, 60-digit Kendall quadrature vs. Dengler (2010) (diff <= 1.4e-51), Richardson 1/nu expansion coefficients (nu = 400, 800), and the crossover bracket.- spearman_null_variance_reproduction.json : Machine-readable JSON output reproducing all numerical checks, tables, Monte Carlo runs, and crossover diagnostics (all_pass: true).- K-PROTOCOL Master Series Catalog and 28-Suite DOI Index.pdf : Updated 28-suite architectural blueprint and permanent Concept DOI directory (Suites 01-28), including Division IV (Mathematical Statistics & Information Geometry).- k_protocol_official_zenodo_repository_downloader_28_suites.py : Official Python repository downloader supporting selective retrieval of Foundational Roots (01-03), All 28 Suites (01-28), or Mathematical Statistics Suites (24-28). Division IV: Mathematical Statistics & Information Geometry

Zenodo (CERN European Organization for Nuclear Research)
Probability and Statistical Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon