A multi-resolution approximation via linear projection for large spatial datasets
From MaRDI portal
Publication:825323
DOI10.1007/S42081-020-00092-XzbMATH Open1477.62267arXiv2004.05102OpenAlexW3097080577MaRDI QIDQ825323FDOQ825323
Authors: Toshihiro Hirano 
Publication date: 17 December 2021
Published in: Japanese Journal of Statistics and Data Science (Search for Journal in Brave)
Abstract: Recent technical advances in collecting spatial data have been increasing the demand for methods to analyze large spatial datasets. The statistical analysis for these types of datasets can provide useful knowledge in various fields. However, conventional spatial statistical methods, such as maximum likelihood estimation and kriging, are impractically time-consuming for large spatial datasets due to the necessary matrix inversions. To cope with this problem, we propose a multi-resolution approximation via linear projection (-RA-lp). The -RA-lp conducts a linear projection approach on each subregion whenever a spatial domain is subdivided, which leads to an approximated covariance function capturing both the large- and small-scale spatial variations. Moreover, we elicit the algorithms for fast computation of the log-likelihood function and predictive distribution with the approximated covariance function obtained by the -RA-lp. Simulation studies and a real data analysis for air dose rates demonstrate that our proposed -RA-lp works well relative to the related existing methods.
Full work available at URL: https://arxiv.org/abs/2004.05102
Recommendations
- A class of multi-resolution approximations for large spatial datasets
- A Full Scale Approximation of Covariance Functions for Large Spatial Data Sets
- Kryging: geostatistical analysis of large-scale datasets using Krylov subspace methods
- Modified linear projection for large spatial datasets
- Smoothed full-scale approximation of Gaussian process models for computation of large spatial data sets
geostatisticsGaussian processcovariance taperinglarge spatial datasetsmulti-resolution approximationstochastic matrix approximation
Cites Work
- Fixed Rank Kriging for Very Large Spatial Data Sets
- Strictly Proper Scoring Rules, Prediction, and Estimation
- Fixed-domain asymptotic properties of tapered maximum likelihood estimators
- Estimation and prediction using generalized Wendland covariance functions under fixed domain asymptotics
- Compactly supported correlation functions
- Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree
- On fixed-domain asymptotics and covariance tapering in Gaussian random field models
- A case study competition among methods for analyzing large spatial data
- Statistics for spatio-temporal data
- Gaussian Predictive Process Models for Large Spatial Data Sets
- Approximating Likelihoods for Large Spatial Data Sets
- Approximate Likelihood for Large Irregularly Spaced Spatial Data
- Covariance tapering for likelihood-based estimation in large spatial data sets
- A general framework for Vecchia approximations of Gaussian processes
- An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach
- A Full Scale Approximation of Covariance Functions for Large Spatial Data Sets
- Title not available (Why is that?)
- A class of multi-resolution approximations for large spatial datasets
- Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
- Improving the performance of predictive process modeling for large datasets
- Statistics for spatial data
- Title not available (Why is that?)
- Efficient Gaussian process regression for large datasets
- Fourier analysis of irregularly spaced data on \(R^d\)
- Title not available (Why is that?)
- Matrix tricks for linear statistical models. Our personal top twenty.
- On the Early History of the Singular Value Decomposition
- Asymptotic properties of multivariate tapering for estimation and prediction
- Spectral density estimation for random fields via periodic embeddings
- Covariance tapering for prediction of large spatial data sets in transformed random fields
- Estimating Extremal Eigenvalues and Condition Numbers of Matrices
- Penalized maximum likelihood estimation and variable selection in geostatistics
- Locally stationary spatio-temporal processes
- Modified linear projection for large spatial datasets
- Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data
Cited In (6)
- A Fast, Optimal Spatial-Prediction Method for Massive Datasets
- Modified linear projection for large spatial datasets
- Title not available (Why is that?)
- A class of multi-resolution approximations for large spatial datasets
- Multi-scale shotgun stochastic search for large spatial datasets
- Kryging: geostatistical analysis of large-scale datasets using Krylov subspace methods
This page was built for publication: A multi-resolution approximation via linear projection for large spatial datasets
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q825323)
