Neighborhood radius estimation for variable-neighborhood random fields

DOI10.1016/J.SPA.2011.05.001zbMATH Open1398.62058arXiv1002.4850OpenAlexW2058049285WikidataQ98839680 ScholiaQ98839680MaRDI QIDQ554466FDOQ554466

Authors: E. Löcherbach, Enza Orlandi 

Publication date: 4 August 2011

Published in: Stochastic Processes and their Applications (Search for Journal in Brave)

Abstract: We consider random fields defined by finite-region conditional probabilities depending on a neighborhood of the region which changes with the boundary conditions. To predict the symbols within any finite region it is necessary to inspect a random number of neighborhood symbols which might change according to the value of them. In analogy to the one dimensional setting we call these neighborhood symbols the context of the region. This framework is a natural extension, to d-dimensional fields, of the notion of variable-length Markov chains introduced by Rissanen (1983) in his classical paper. We define an algorithm to estimate the radius of the smallest ball containing the context based on a realization of the field. We prove the consistency of this estimator. Our proofs are constructive and yield explicit upper bounds for the probability of wrong estimation of the radius of the context.

Full work available at URL: https://arxiv.org/abs/1002.4850

Recommendations

• Estimate of the interaction neighborhood radius for a Markov random field
• Consistent estimation of the basic neighborhood of Markov random fields
• Partially observed Markov random fields are variable neighborhood random fields
• Exponential inequalities for VLMC empirical trees
• An oracle approach for interaction neighborhood estimation in random fields

zbMATH Keywords

consistent estimationGibbs measuresrandom lattice fieldscontext algorithmvariable-neighborhood random fields

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Random fields; image analysis (62M40) Random fields (60G60) Geometric probability and stochastic geometry (60D05)

Cites Work

• Title not available (Why is that?)
• Context tree estimation for not necessarily finite memory processes, via BIC and MDL
• Infinite systems of interacting chains with memory of variable length -- a stochastic model for biological neural nets
• Exponential inequalities and functional central limit theorems for random fields
• Gibbs measures and phase transitions
• The problem of uniqueness of a Gibbsian random field and the problem of phase transitions
• Practical simulation and estimation for Gibbs Delaunay-Voronoi tessellations with geometric hardcore interaction
• Refinements of Pinsker's inequality
• Existence of Gibbsian point processes with geometry-dependent interactions
• Scaling limits in statistical mechanics and microstructures in continuum mechanics
• Gibbsian random fields for lattice systems with pairwise interactions.
• On Bernstein-type inequalities for martingales.
• On consistency of a class of estimators for exponential families of Markov random fields on the lattice
• Exponential inequalities for empirical unbounded context trees
• A universal data compression system
• The optimal error exponent for Markov order estimation
• Exponential inequalities for VLMC empirical trees
• A consistent model selection procedure for Markov random fields based on penalized pseudolikelihood
• Consistent estimation of the basic neighborhood of Markov random fields
• Estimation of General Stationary Processes by Variable Length Markov Chains
• On the estimation of the order of a Markov chain and universal data compression

Cited In (10)

• Asymptotic normality of the maximum likelihood estimator for cooperative sequential adsorp\-tion
• Identifying interacting pairs of sites in Ising models on a countable set
• On universal algorithms for classifying and predicting stationary processes
• Data-driven neighborhood selection of a Gaussian field
• Estimate of the interaction neighborhood radius for a Markov random field
• Structure recovery for partially observed discrete Markov random fields on graphs under not necessarily positive distributions
• Model selection for Markov random fields on graphs under a mixing condition
• Almost Gibbsianness and parsimonious description of the decimated 2d-Ising model
• Partially observed Markov random fields are variable neighborhood random fields
• Estimating the interaction graph of stochastic neural dynamics

This page was built for publication: Neighborhood radius estimation for variable-neighborhood random fields

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q554466)