Saved in:
Bibliographic Details
Main Authors: Derumigny, Alexis, Schmidt-Hieber, Johannes
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2303.08122
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909196399673344
author Derumigny, Alexis
Schmidt-Hieber, Johannes
author_facet Derumigny, Alexis
Schmidt-Hieber, Johannes
contents We propose a new concept of codivergence, which quantifies the similarity between two probability measures $P_1, P_2$ relative to a reference probability measure $P_0$. In the neighborhood of the reference measure $P_0$, a codivergence behaves like an inner product between the measures $P_1 - P_0$ and $P_2 - P_0$. Codivergences of covariance-type and correlation-type are introduced and studied with a focus on two specific correlation-type codivergences, the $χ^2$-codivergence and the Hellinger codivergence. We derive explicit expressions for several common parametric families of probability distributions. For a codivergence, we introduce moreover the divergence matrix as an analogue of the Gram matrix. It is shown that the $χ^2$-divergence matrix satisfies a data-processing inequality.
format Preprint
id arxiv_https___arxiv_org_abs_2303_08122
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Codivergences and information matrices
Derumigny, Alexis
Schmidt-Hieber, Johannes
Statistics Theory
Information Theory
Probability
62B11, 46E27, 15A63
We propose a new concept of codivergence, which quantifies the similarity between two probability measures $P_1, P_2$ relative to a reference probability measure $P_0$. In the neighborhood of the reference measure $P_0$, a codivergence behaves like an inner product between the measures $P_1 - P_0$ and $P_2 - P_0$. Codivergences of covariance-type and correlation-type are introduced and studied with a focus on two specific correlation-type codivergences, the $χ^2$-codivergence and the Hellinger codivergence. We derive explicit expressions for several common parametric families of probability distributions. For a codivergence, we introduce moreover the divergence matrix as an analogue of the Gram matrix. It is shown that the $χ^2$-divergence matrix satisfies a data-processing inequality.
title Codivergences and information matrices
topic Statistics Theory
Information Theory
Probability
62B11, 46E27, 15A63
url https://arxiv.org/abs/2303.08122