Data-driven neighborhood selection of a Gaussian field
From MaRDI portal
Publication:962389
Abstract: We study the nonparametric covariance estimation of a stationary Gaussian field X observed on a lattice. To tackle this issue, a neighborhood selection procedure has been recently introduced. This procedure amounts to selecting a neighborhood m by a penalization method and estimating the covariance of X in the space of Gaussian Markov random fields (GMRFs) with neighborhood m. Such a strategy is shown to satisfy oracle inequalities as well as minimax adaptive properties. However, it suffers several drawbacks which make the method difficult to apply in practice. On the one hand, the penalty depends on some unknown quantities. On the other hand, the procedure is only defined for toroidal lattices. The present contribution is threefold. A data-driven algorithm is proposed for tuning the penalty function. Moreover, the procedure is extended to non-toroidal lattices. Finally, numerical study illustrate the performances of the method on simulated examples. These simulations suggest that Gaussian Markov random field selection is often a good alternative to variogram estimation.
Recommendations
- scientific article; zbMATH DE number 78500
- Optimization of Gaussian random fields
- Selection of the neighborhood structure for space-time Markov random field models
- Nearest neighbor estimators for random fields
- Neighborhood radius estimation for variable-neighborhood random fields
- Region selection in Markov random fields: Gaussian case
- Bayesian selection of the neighbourhood order for Gauss–Markov texture models
- Gaussian processes and Gaussian random fields
Cites work
- scientific article; zbMATH DE number 954974 (Why is no real title available?)
- scientific article; zbMATH DE number 1219611 (Why is no real title available?)
- scientific article; zbMATH DE number 1134987 (Why is no real title available?)
- scientific article; zbMATH DE number 3444596 (Why is no real title available?)
- scientific article; zbMATH DE number 783366 (Why is no real title available?)
- A comparative study of Gaussian geostatistical models and Gaussian Markov random field models
- Adaptive estimation of a quadratic functional by model selection.
- Adaptive estimation of stationary Gaussian fields
- Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations (with discussion)
- Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
- Conditional-mean least-squares fitting of Gaussian Markov random fields to Gaussian fields
- Detecting multiple change-points in the mean of Gaussian process by model selection
- Edge Detection, Spatial Smoothing, and Image Reconstruction With Partially Observed Multivariate Data
- Estimating the dimension of a model
- Fitting Gaussian Markov Random Fields to Gaussian Fields
- Gaussian Markov Random Fields
- Image denoising using scale mixtures of gaussians in the wavelet domain
- Interpolation of spatial data. Some theory for kriging
- Minimal penalties for Gaussian model selection
- On the nonparametric estimation of covariance functions
- On the underfitting and overfitting sets of models chosen by order selection criteria.
- Selecting models focussing on the modeller's purpose
- Semiparametric Estimation of Spectral Density With Irregular Observations
- Semiparametric estimation of spatial long-range dependence
- Spatial variation. 2nd ed
- Statistics for spatial data
- Toeplitz and circulant matrices: a review.
- Valid parameter space of 2-D Gaussian Markov random fields
Cited in
(5)- Slope heuristics: overview and implementation
- A comparison of sampling grids, cut-off distance and type of residuals in parametric variogram estimation
- Adaptive estimation of stationary Gaussian fields
- Region selection in Markov random fields: Gaussian case
- Estimation of covariance functions by a fully data-driven model selection procedure and its application to Kriging spatial interpolation of real rainfall data
This page was built for publication: Data-driven neighborhood selection of a Gaussian field
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q962389)
