Paperlane

no. 13 · Wednesday 11 November 2026

A Note on the Cut-Locus-Corrected Fréchet Mean Central Limit Theorem

Susovan Pal

math.ST · v3

Abstract

We record three observations concerning \cite{HLW24}. First, the proof of its Theorem~1 yields the population linearization (13)--(14) at \textit{any critical point} of the Fréchet function $F_μ$, under its remaining geometric and measure-theoretic assumptions; neither \textit{isolation} nor a \textit{nonsingular} Hessian of $F_μ$ is needed. Second, its Theorem~2 is given a shorter proof by applying the M-estimator theorem \cite[Theorem~5.23, pp.~53–54]{vdV98} to negative squared distance in normal coordinates, as opposed to the Donsker argument used there. Third, the above shorter route also helps us establish a CLT for a measurable selection of strict local empirical Fréchet mean with positive definite Hessian, with the cut locus correction, relaxing the assumptions of uniqueness or global minimality, as in \cite{HLW24}. Numerical experiments at the end show the agreement of the Hessian of the Fréchet function at \textit{local} Fréchet means with the (signed) sum of the expected Hessian of the squared distance function and the cut locus correction there.

fields
Statistics Theory
pdf ↗ page arXiv:2611.07956 ↗