Analytic asymptotic null distribution of the quotient-affine distance

Derive an analytic asymptotic null distribution for the quotient-affine geodesic distance between sample correlation matrices.

Background

The quotient-affine geodesic distance provides an intrinsic global measure of separation between full-rank correlation matrices. Before the paper's local asymptotic argument, an analytic null distribution for this nonlinear distance was not available to the authors.

The paper subsequently derives a Gaussian local asymptotic result by relating the squared geodesic distance to Jennrich's quadratic statistic, showing convergence after effective-sample-size scaling to a chi-squared distribution. The explicitly stated unresolved issue concerns obtaining an analytic asymptotic null distribution for the distance itself, beyond its local connection to Jennrich's statistic.

References

We are also not aware of an analytic asymptotic null distribution for this distance.

Connecting Riemannian Geometry and Statistical Inference for Correlation Matrices  (2608.27209 - Kuketayev, 27 Aug 2026) in Section 1, Introduction