A general framework for Vecchia approximations of Gaussian processes

DOI10.1214/19-sts755OpenAlexW3114272883WikidataQ114060538 ScholiaQ114060538MaRDI QIDQ6032766

Matthias Katzfuss, Joseph Guinness

Publication date: 6 July 2021

Published in: Statistical Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1708.06302

zbMATH Keywords

computational complexityspatial statisticssparsitydirected acyclic graphslarge datasetscovariance approximation

Mathematics Subject Classification ID

Statistics (62-XX)

Related Items (44)

Bayesian inference for high-dimensional nonstationary Gaussian processes ⋮ A multi-resolution approximation via linear projection for large spatial datasets ⋮ Finite Element Representations of Gaussian Processes: Balancing Numerical and Statistical Accuracy ⋮ Bayesian multiresolution modeling of georeferenced data: an extension of `LatticeKrig' ⋮ Making Recursive Bayesian Inference Accessible ⋮ An Approach to Incorporate Subsampling Into a Generic Bayesian Hierarchical Model ⋮ Multi-Resolution Filters for Massive Spatio-Temporal Data ⋮ Large-scale local surrogate modeling of stochastic simulation experiments ⋮ Scaled Vecchia Approximation for Fast Computer-Model Emulation ⋮ Scalable semiparametric spatio-temporal regression for large data analysis ⋮ Bayesian latent variable co-kriging model in remote sensing for quality flagged observations ⋮ Latent multivariate log-gamma models for high-dimensional multitype responses with application to daily fine particulate matter and mortality counts ⋮ Distributed Bayesian inference in massive spatial data ⋮ Bayesian nonstationary and nonparametric covariance estimation for large spatial data (with discussion) ⋮ Gaussian orthogonal latent factor processes for large incomplete matrices of correlated data ⋮ Bayesian hierarchical modeling and analysis for actigraph data from wearable devices ⋮ Correlated product of experts for sparse Gaussian process regression ⋮ Fast Bayesian inference of block nearest neighbor Gaussian models for large data ⋮ Correlation-based sparse inverse Cholesky factorization for fast Gaussian-process inference ⋮ Scalable computations for nonstationary Gaussian processes ⋮ Nearest-neighbor mixture models for non-Gaussian spatial processes ⋮ Highly Scalable Bayesian Geostatistical Modeling via Meshed Gaussian Processes on Partitioned Domains ⋮ Unnamed Item ⋮ Scalable penalized spatiotemporal land-use regression for ground-level nitrogen dioxide ⋮ Unifying compactly supported and Matérn covariance functions in spatial statistics ⋮ Gaussian Process Learning via Fisher Scoring of Vecchia's Approximation ⋮ High-dimensional order-free multivariate spatial disease mapping ⋮ Vecchia-approximated Deep Gaussian Processes for Computer Experiments ⋮ Covariance Models and Simulation Algorithm for Stationary Vector Random Fields on Spheres Crossed with Euclidean Spaces ⋮ Multivariate isotropic random fields on spheres: nonparametric Bayesian modeling and \(L^p\) fast approximations ⋮ Max-and-smooth: a two-step approach for approximate Bayesian inference in latent Gaussian models ⋮ Nonparametric spectral methods for multivariate spatial and spatial-temporal data ⋮ Locally induced Gaussian processes for large-scale simulation experiments ⋮ Scalable Gaussian Process Computations Using Hierarchical Matrices ⋮ Towards a complete picture of stationary covariance functions on spheres cross time ⋮ A hierarchical multivariate spatio-temporal model for clustered climate data with annual cycles ⋮ Regularity and approximation of Gaussian random fields evolving temporally over compact two-point homogeneous spaces ⋮ Improving performances of MCMC for nearest neighbor Gaussian process models with full data augmentation ⋮ Mixture of multivariate Gaussian processes for classification of irregularly sampled satellite image time-series ⋮ Hierarchical Bayesian modeling of ocean heat content and its uncertainty ⋮ Sparse Cholesky Factorization by Kullback--Leibler Minimization ⋮ Discussion on competition on spatial statistics for large datasets ⋮ Multi-scale Vecchia approximations of Gaussian processes ⋮ Hierarchical sparse Cholesky decomposition with applications to high-dimensional spatio-temporal filtering

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• An Explicit Link between Gaussian Fields and Gaussian Markov Random Fields: The Stochastic Partial Differential Equation Approach
• Efficient Algorithms for Bayesian Nearest Neighbor Gaussian Processes
• 2010 Rietz lecture: When does the screening effect hold?
• Cases for the nugget in modeling computer experiments
• Covariance approximation for large multivariate spatial data sets with an application to multiple climate model errors
• Improving the performance of predictive process modeling for large datasets
• Fixed-domain asymptotic properties of tapered maximum likelihood estimators
• Image analysis with partially ordered Markov models.
• High dimensional sparse covariance estimation via directed acyclic graphs
• A case study competition among methods for analyzing large spatial data
• Nonseparable dynamic nearest neighbor Gaussian process models for large spatio-temporal data with an application to particulate matter analysis
• Hierarchical Low Rank Approximation of Likelihoods for Large Spatial Datasets
• Fixed Rank Kriging for Very Large Spatial Data Sets
• Gaussian Predictive Process Models for Large Spatial Data Sets
• Fast Spatial Gaussian Process Maximum Likelihood Estimation via Skeletonization Factorizations
• A Full Scale Approximation of Covariance Functions for Large Spatial Data Sets
• Approximating Likelihoods for Large Spatial Data Sets
• Gaussian Markov Random Fields
• A dimension-reduced approach to space-time Kalman filtering
• A class of multi-resolution approximations for large spatial datasets
• Smoothed Full-Scale Approximation of Gaussian Process Models for Computation of Large Spatial Datasets
• Covariance Tapering for Likelihood-Based Estimation in Large Spatial Data Sets
• Spatio‐temporal smoothing and EM estimation for massive remote‐sensing data sets

This page was built for publication: A general framework for Vecchia approximations of Gaussian processes