Differential geometry of curved exponential families. Curvatures and information loss

DOI10.1214/AOS/1176345779zbMATH Open0507.62026OpenAlexW2025994725WikidataQ115240732 ScholiaQ115240732MaRDI QIDQ1837486FDOQ1837486

Authors: Shun-Ichi Amari 

Publication date: 1982

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1214/aos/1176345779

zbMATH Keywords

Kullback-Leibler distancedualitycurved exponential familiesmaximum-likelihood estimatorLegendre transformationrecovery of informationstatistical curvaturesFisher-efficient estimatorsmulti-parameter families of distributionssecond- order efficiencysecond-order information lossstatistical affine connections

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Asymptotic properties of parametric estimators (62F12) Asymptotic distribution theory in statistics (62E20) Linear and affine connections (53B05) Classical differential geometry (53A99)

Cited In (only showing first 100 items - show all)

• Statistical information approaches for the modelling of the epileptic brain
• Geometry of exponential type regression models and asymptotic inference
• Differential geometry of a parametric family of invertible linear systems—Riemannian metric, dual affine connections, and divergence
• Reference priors via \(\alpha \)-divergence for a certain non-regular model in the presence of a nuisance parameter
• Shrinkage and modification techniques in estimation of variance and the related problems: A review
• On divergence measures leading to Jeffreys and other reference priors
• A class of tests for a general covariance structure
• Estimation, prediction and the Stein phenomenon under divergence loss
• Divergence Function, Duality, and Convex Analysis
• On the Stein phenomenon under divergence loss and an unknown variance-covariance matrix
• On the best equivariant estimator of mean of a multivariate normal population
• On the geometric structure ofsome statistical manifolds
• Objective priors: an introduction for frequentists
• Expert information and nonparametric Bayesian inference of rare events
• Some asymptotic inference on nonlinear models with random effects (a geometric approach)
• Nonparametric information geometry: from divergence function to referential-representational biduality on statistical manifolds
• New results on mixture and exponential models by Orlicz spaces
• Exponential statistical manifold
• Reparametrizations and metric structures in thermodynamic phase space
• A geometric approach of the generalized least-squares estimation in analysis of covariance structures
• On the geometry of generalized Gaussian distributions
• A general divergence criterion for prior selection
• Density prediction and the Stein phenomenon
• Invited discussion paper small-sample distributional properties of nonlinear regression estimators (a geometric approach)
• Information geometric methods for complexity
• Information theory for maximum likelihood estimation of diffusion models
• Asymptotical improvement of maximum likelihood estimators on Kullback-Leibler loss
• Some asymptotic inference in multinomial nonlinear models (a geometric approach)
• \(F\)-geometry and Amari's \(\alpha\)-geometry on a statistical manifold
• Gradient systems in view of information geometry
• The relation between information theory and the differential geometry approach to statistics
• Geometry of the log-likelihood ratio statistic in misspecified models
• The maximum likelihood prior
• DIFFERENTIAL GEOMETRY OF ARMA MODELS
• Mutual information, metric entropy and cumulative relative entropy risk
• Exponential and mixture families in quantum statistics: Dual structure and unbiased parameter estimation
• Differential geometry of Edgeworth expansions in curved exponential family
• Discussion on ``Objective priors: an introduction for frequentists by M. Ghosh
• Computer geometry and encoding the information on a manifold
• The geometry of case deletion and the assessment of influence in nonlinear regression
• A geometry on the space of probabilities. II: Projective spaces and exponential families
• On preferred point geometry in statistics
• Detecting higher-order interactions among the spiking events in a group of neurons
• Distribution of directional data and fabric tensors
• Statistical manifolds with almost quaternionic structures and quaternionic Kähler-like statistical submersions
• Statistical manifold as an affine space: a functional equation approach
• Inference on the eigenvalues of the covariance matrix of a multivariate normal distribution -- geometrical view
• Information geometry and sufficient statistics
• Statistical morphisms and related invariance properties
• Fisher information under restriction of Shannon information in multi-terminal situations
• Best equivariant estimation in curved covariance models
• Recursive nonlinear estimation: Geometry of a space of posterior densities
• A characterization of extended monotone metrics
• NEW LARGE SAMPLE AND BOOTSTRAP METHODS ON SHAPE SPACES IN HIGH LEVEL ANALYSIS OF NATURAL IMAGES
• The impact of high-order interactions on the rate of synchronous discharge and information transmission in somatosensory cortex
• On confidence regions of semiparametric nonlinear reproductive dispersion models
• Connections on Non-Parametric Statistical Manifolds by Orlicz Space Geometry
• A new approach to portfolio theory
• A characterization of second order efficiency in a curved exponential family
• Equivariant estimation under the pitman closeness criterion
• On the best equivariant estimator of covariance matrix of a multivariate normal population
• Information Geometry
• A note on the interpretation of the Bahadur bound and the rate of convergence of the maximum likelihood estimator
• Equivariant estimation of a mean vector \(\mu\) of N(\(\mu\) ,\(\Sigma\) ) with \(\mu '\Sigma ^{-1}\mu =1\) or \(\Sigma ^{-}\mu =c\) or \(\Sigma =\sigma\) 2\(\mu\) '\(\mu\) I
• The algebro-geometric study of range maps
• Variational representations of annealing paths: Bregman information under monotonic embedding
• Differential analytic aspects of statistical models
• Affine statistical bundle modeled on a Gaussian Orlicz-Sobolev space
• Half a century of information geometry, part 1
• Gram matrices of quantum channels via quantum Fisher information with applications to decoherence and uncertainty
• Minimum information divergence of Q-functions for dynamic treatment resumes
• Legendre duality: from thermodynamics to information geometry
• Discrete mixture representations of parametric distribution families: geometry and statistics
• Renormalization group flows from the Hessian geometry of quantum effective actions
• Asymptotic properties of conditional maximum likelihood estimator in a certain exponential model
• On the Loss Robustness of Least-Square Estimators
• A geometric approach to singularity for Hilbert space-valued SDEs
• On the geometrical representation of classical statistical mechanics
• Informative fractal dimension associated with nonmetricity in information geometry
• Geometrical properties of Pareto distribution.
• Nonlinear regression and curved exponential families. Improvement of the approximation to the asymptotic distribution
• Curvature and inference for maximum likelihood estimates
• Lightlike hypersurfaces of an indefinite quasi-Sasakian statistical manifold
• Conditions for the existence of a generalization of Rényi divergence
• MULTIMODALITY p**-FORMULA AND CONFIDENCE REGIONS
• On a criterion of Riemannian distance for singularity and absolute continuity of probability measures.
• The Information Geometric Structure of Generalized Empirical Likelihood Estimators
• Title not available (Why is that?)
• Conformal vector fields on statistical manifolds
• Quasi-likelihood or extended quasi-likelihood? An information-geometric approach
• Quantum and Fisher information from the Husimi and related distributions
• Information geometry on the curved \(q\)-exponential family with application to survival data analysis
• Estimating Spiking Irregularities Under Changing Environments
• Information theory unification of epidemiological and population dynamics
• On lightlike geometry of indefinite Sasakian statistical manifolds
• Geometry of parametric binary choice models
• Information geometric optimal reconfiguration of pulse-Doppler sensor parameters
• Lifting statistical structures
• Gram-Schmidt-Fisher scoring algorithm for parameter orthogonalization in MLE
• Natural alpha embeddings

This page was built for publication: Differential geometry of curved exponential families. Curvatures and information loss

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1837486)