Abstract:The (CLT) central limit theorems for generalized Frechet means (data descriptors assuming values in stratified spaces, such as intrinsic means, geodesics, etc.) on manifolds from the literature are only valid if a certain empirical process of Hessians of the Frechet function converges suitably, as in the proof of the prototypical BP-CLT (Bhattacharya and Patrangenaru (2005)). This is not valid in many realistic scenarios and we provide for a new very general CLT. In particular this includes scenarios where, in a suitable chart, the sample mean fluctuates asymptotically at a scale $n^{\alpha}$ with exponents ${\alpha} < 1/2$ with a non-normal distribution. As the BP-CLT yields only fluctuations that are, rescaled with $n^{1/2}$ , asymptotically normal, just as the classical CLT for random vectors, these lower rates, somewhat loosely called smeariness, had to date been observed only on the circle (Hotz and Huckemann (2015)). We make the concept of smeariness on manifolds precise, give an example for two-smeariness on spheres of arbitrary dimension, and show that smeariness, although "almost never" occurring, may have serious statistical implications on a continuum of sample scenarios nearby. In fact, this effect increases with dimension, striking in particular in high dimension low sample size scenarios.

Central limit theorem for components in meandric systems through high moments

On Central Limit Theorems for Vector Random Measures and Measure-Valued Processes

Three central limit theorems for the unbounded excursion component of a Gaussian field

Functional Central Limit Theorems for Constrained Mittag-Leffler Ensemble in Hard Edge Scaling

Lindeberg's Central Limit Theorems for Martingale-Like Sequences under Sub-Linear Expectations

Central limits from generating functions

Central limit theorem for smooth statistics of one-dimensional free fermions

Multivariate CLT for critical points

Central limit theorems for the shrinking target problem

Generalization of the Central Limit Theorem to Critical Systems: Revisiting Perturbation Theory

Multivariate asymptotic normality determined by high moments

Central limit theorems with a rate of convergence for sequences of transformations

Central limit theorem for random partitions under the Plancherel measure

The central limit theorem for sum-functions of m-tuples of spacings

A Smeary Central Limit Theorem for Manifolds with Application to High Dimensional Spheres

Central limit theorem on CAT(0) spaces with contracting isometries

A Central Limit Theorem for Modified Massive Arratia Flow

Asymptotic Multiplicity Distributions and Analog of the Central-Limit Theorem

Central Limit Theorem in Disordered Monomer-Dimer Model

On the Almost Sure Central Limit Theorem for Vector Martingales: Convergence of Moments and Statistical Applications

Local central limit theorem for gradient field models