The Poisson tensor completion parametric estimator

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Dunlavy, Daniel M., Lehoucq, Richard B., Mayer, Carolyn D., Prasadan, Arvind
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911494552158208
author Dunlavy, Daniel M.
Lehoucq, Richard B.
Mayer, Carolyn D.
Prasadan, Arvind
author_facet Dunlavy, Daniel M.
Lehoucq, Richard B.
Mayer, Carolyn D.
Prasadan, Arvind
contents We introduce the Poisson tensor completion (PTC) estimator that exploits inter-sample relationships to compute a low-rank Poisson tensor decomposition of the frequency histogram for samples of a multivariate distribution. Our crucial observation is that the histogram bins are an instance of a space partitioning of counts and thus can be identified with a spatial non-homogeneous Poisson process. The Poisson tensor decomposition leads to a completion of the mean measure over all bins -- including those containing few to no samples -- and leads to our proposed estimator. A Poisson tensor decomposition models the underlying distribution of the count data and guarantees non-negative estimated values obviating the need for additional constraints to ensure non-negativity. Furthermore, we demonstrate that our PTC estimator is a substantial improvement over standard histogram-based estimators for sub-Gaussian probability distributions because of the concentration of norm phenomenon.
format Preprint
id arxiv_https___arxiv_org_abs_2505_04957
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Poisson tensor completion parametric estimator
Dunlavy, Daniel M.
Lehoucq, Richard B.
Mayer, Carolyn D.
Prasadan, Arvind
Statistics Theory
Methodology
We introduce the Poisson tensor completion (PTC) estimator that exploits inter-sample relationships to compute a low-rank Poisson tensor decomposition of the frequency histogram for samples of a multivariate distribution. Our crucial observation is that the histogram bins are an instance of a space partitioning of counts and thus can be identified with a spatial non-homogeneous Poisson process. The Poisson tensor decomposition leads to a completion of the mean measure over all bins -- including those containing few to no samples -- and leads to our proposed estimator. A Poisson tensor decomposition models the underlying distribution of the count data and guarantees non-negative estimated values obviating the need for additional constraints to ensure non-negativity. Furthermore, we demonstrate that our PTC estimator is a substantial improvement over standard histogram-based estimators for sub-Gaussian probability distributions because of the concentration of norm phenomenon.
title The Poisson tensor completion parametric estimator
topic Statistics Theory
Methodology
url https://arxiv.org/abs/2505.04957