Manifolds of differentiable densities

DOI10.1051/PS/2018003zbMATH Open1410.46056arXiv1608.03979OpenAlexW2963628670MaRDI QIDQ4615428FDOQ4615428

Authors: Nigel J. Newton 

Publication date: 28 January 2019

Published in: ESAIM: Probability and Statistics (Search for Journal in Brave)

Abstract: We develop a family of infinite-dimensional (non-parametric) manifolds of probability measures. The latter are defined on underlying Banach spaces, and have densities of class $C_b^k$ with respect to appropriate reference measures. The case $k=infty$, in which the manifolds are modelled on Fr'{e}chet spaces, is included. The manifolds admit the Fisher-Rao metric and, unusually for the non-parametric setting, Amari's $alpha$-covariant derivatives for all $alphainR$. By construction, they are $C^{i}nfty$-embedded submanifolds of particular manifolds of finite measures. The statistical manifolds are dually ($alpha=pm1$) flat, and admit mixture and exponential representations as charts. Their curvatures with respect to the $alpha$-covariant derivatives are derived. The likelihood function associated with a finite sample is a continuous function on each of the manifolds, and the $alpha$-divergences are of class $C^{i}nfty$.

Full work available at URL: https://arxiv.org/abs/1608.03979

Recommendations

• Infinite-dimensional statistical manifolds based on a balanced chart
• Infinite-dimensional manifolds of finite-entropy probability measures
• A class of non-parametric statistical manifolds modelled on Sobolev space
• Nonparametric information geometry
• An infinite-dimensional statistical manifold modelled on Hilbert space

zbMATH Keywords

Banach manifoldinformation geometryFisher-Rao metricnon-parametric statisticsFréchet manifold

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Nonparametric estimation (62G05) Spaces of measures (46E27) Applications of functional analysis in probability theory and statistics (46N30) Geometric probability and stochastic geometry (60D05) Measures of information, entropy (94A17) Infinite-dimensional manifolds (46T05)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• The exponential statistical manifold: mean parameters, orthogonality and space transformations
• A infinite-dimensional geometric structure on the space of all the probability measures equivalent to a given one
• Infinite-dimensional statistical manifolds based on a balanced chart
• Generalised Thermostatistics
• Geometric science of information. First international conference, GSI 2013, Paris, France, August 28--30, 2013. Proceedings
• Title not available (Why is that?)
• Connections on Non-Parametric Statistical Manifolds by Orlicz Space Geometry
• An infinite-dimensional statistical manifold modelled on Hilbert space
• Title not available (Why is that?)
• Title not available (Why is that?)
• Exponential statistical manifold
• Geometric science of information. Second international conference, GSI 2015, Palaiseau, France, October 28--30, 2015. Proceedings
• Information geometry and sufficient statistics
• Information geometry formalism for the spatially homogeneous Boltzmann equation
• On a differential structure for the group of diffeomorphisms
• Quantum probability for probabilists.
• Geometry in a Fréchet context. A projective limit approach
• Parametrized measure models
• A \(q\)-exponential statistical Banach manifold
• Uniqueness of the Fisher-Rao metric on the space of smooth densities
• Information geometric nonlinear filtering

Cited In (14)

• Global differentiable structures for the Fisher-Rao and Kantorovich-Wasserstein-Otto metrics
• Natural differentiable structures on statistical models and the Fisher metric
• Infinite-dimensional statistical manifolds based on a balanced chart
• An infinite-dimensional statistical manifold modelled on Hilbert space
• Dual connections in nonparametric classical information geometry
• Infinite-dimensional manifolds of finite-entropy probability measures
• Asymmetric topologies on statistical manifolds
• Title not available (Why is that?)
• A class of non-parametric statistical manifolds modelled on Sobolev space
• Geometry of manifolds with densities
• Statistical manifold as an affine space: a functional equation approach
• Parametrized measure models
• Geometry of the Fisher-Rao metric on the space of smooth densities on a compact manifold
• Title not available (Why is that?)

This page was built for publication: Manifolds of differentiable densities

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4615428)