English

On a completeness problem in a Fourier-based probability metrics in $\mathbb{R}^N$

Probability 2019-09-30 v4 Classical Analysis and ODEs

Abstract

We study completeness of the spaces Ps=\mathcal{P}_s^= of probability measures in RN\mathbb{R}^N which have equal (prescribed) moments up to order sNs \in \mathbb{N}, endowed with the metric ds(μ,ν)=supxRN0μ^(x)ν^(x)xsd_s(\mu,\nu)=\sup_{x \in \mathbb{R}^N\setminus 0}\frac{|\hat \mu(x)-\hat \nu(x)|}{|x|^s}, where μ^\hat \mu is the characteristic function of μ\mu. We prove that the spaces (Ps=,ds)(\mathcal{P}_s^=,d_s) are complete if ss is even and construct suitable counterexamples to completeness for all odd ss. This solves an open problem formulated by J. Carrillo and G. Toscani in 2007.

Keywords

Cite

@article{arxiv.1609.00343,
  title  = {On a completeness problem in a Fourier-based probability metrics in $\mathbb{R}^N$},
  author = {Małgorzata Stawiska},
  journal= {arXiv preprint arXiv:1609.00343},
  year   = {2019}
}

Comments

Substantially revised and expanded; new title; updated references

R2 v1 2026-06-22T15:37:57.215Z