Spectral equivalence of normal-scores and Kendall rank inputs

From papers

Let S^n\widehat{\bm{S}}_n be the normal-scores covariance and let Σ^τ\widehat{\bm{\Sigma}}^{\tau} be the sin-transformed Kendall matrix. Assume p/nγp/n\to\gamma. Spectral equivalence of rank inputs. The two matrices have the same limiting empirical spectral distribution, even though

\left\\|\widehat{\bm{S}}_n-\widehat{\bm{\Sigma}}^{\tau}\right\\|_{\mathrm{op}}

does not vanish. A proof would need controls beyond the operator-norm and normalized Frobenius estimates currently used, which the paper's simulations indicate are insufficient for establishing the claimed spectral equivalence.

Progress summary

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Sources & referencesView supporting material

Primary source

Hamid Karamikabir and Mohammad Arashi, “Mens: Nonlinear shrinkage estimation in nonparanormal models for financial applications”, arXiv:2607.19825 (2026).

Solutions 0

No solutions have been posted yet.