Hierarchical sparse Cholesky decomposition with applications to high-dimensional spatio-temporal filtering
From MaRDI portal
Publication:2114043
DOI10.1007/S11222-021-10077-9zbMATH Open1481.62004arXiv2006.16901OpenAlexW3039502708MaRDI QIDQ2114043FDOQ2114043
Marcin Jurek, Matthias Katzfuss
Publication date: 14 March 2022
Published in: Statistics and Computing (Search for Journal in Brave)
Abstract: Spatial statistics often involves Cholesky decomposition of covariance matrices. To ensure scalability to high dimensions, several recent approximations have assumed a sparse Cholesky factor of the precision matrix. We propose a hierarchical Vecchia approximation, whose conditional-independence assumptions imply sparsity in the Cholesky factors of both the precision and the covariance matrix. This remarkable property is crucial for applications to high-dimensional spatio-temporal filtering. We present a fast and simple algorithm to compute our hierarchical Vecchia approximation, and we provide extensions to non-linear data assimilation with non-Gaussian data based on the Laplace approximation. In several numerical comparisons, including a filtering analysis of satellite data, our methods strongly outperformed alternative approaches.
Full work available at URL: https://arxiv.org/abs/2006.16901
state-space modeldata assimilationhierarchical matrixincomplete Cholesky decompositionspatiotemporal statisticsVecchia approximation
Computational methods for problems pertaining to statistics (62-08) Directional data; spatial statistics (62H11)
Cites Work
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- A case study competition among methods for analyzing large spatial data
- Bayesian calibration of computer models. (With discussion)
- Title not available (Why is that?)
- Gaussian Predictive Process Models for Large Spatial Data Sets
- Title not available (Why is that?)
- Approximating Likelihoods for Large Spatial Data Sets
- Title not available (Why is that?)
- A general framework for Vecchia approximations of Gaussian processes
- Handbook of Spatial Statistics
- Accurate Approximations for Posterior Moments and Marginal Densities
- Title not available (Why is that?)
- 2010 Rietz lecture: When does the screening effect hold?
- A class of multi-resolution approximations for large spatial datasets
- Efficient emulators of computer experiments using compactly supported correlation functions, with an application to cosmology
- Compression, inversion, and approximate PCA of dense kernel matrices at near-linear computational complexity
- Improving the performance of predictive process modeling for large datasets
- Statistics for Spatial Data
- Scalable Gaussian Process Computations Using Hierarchical Matrices
- Spatio‐temporal smoothing and EM estimation for massive remote‐sensing data sets
- Title not available (Why is that?)
- Title not available (Why is that?)
- A dimension-reduced approach to space-time Kalman filtering
- Data Assimilation
- Hierarchical Matrices: Algorithms and Analysis
- The screening effect in kriging
- Parallel inference for massive distributed spatial data using low-rank models
- Practical likelihood analysis for spatial generalized linear mixed models
- Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data
- Understanding the Ensemble Kalman Filter
- Stochastic Partial Differential Equation Based Modelling of Large Space–Time Data Sets
- Sparse Cholesky Factorization by Kullback--Leibler Minimization
- Multi-Resolution Filters for Massive Spatio-Temporal Data
- Ensemble Kalman Methods for High-Dimensional Hierarchical Dynamic Space-Time Models
- Fast Kalman filter using hierarchical matrices and a low-rank perturbative approach
Cited In (4)
- Spatial-temporal nonlinear filtering based on hierarchical statistical models
- Title not available (Why is that?)
- Scalable spatio-temporal smoothing via hierarchical sparse Cholesky decomposition
- Spatiotemporal event studies for environmental data under cross-sectional dependence: an application to air quality assessment in Lombardy
Uses Software
This page was built for publication: Hierarchical sparse Cholesky decomposition with applications to high-dimensional spatio-temporal filtering
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2114043)
