Covariance approximation for large multivariate spatial data sets with an application to multiple climate model errors
From MaRDI portal
Publication:765994
DOI10.1214/11-AOAS478zbMATH Open1234.62071arXiv1203.0133OpenAlexW2070647937MaRDI QIDQ765994FDOQ765994
Authors: Huiyan Sang, Mikyoung Jun, Jianhua Z. Huang 
Publication date: 22 March 2012
Published in: The Annals of Applied Statistics (Search for Journal in Brave)
Abstract: This paper investigates the cross-correlations across multiple climate model errors. We build a Bayesian hierarchical model that accounts for the spatial dependence of individual models as well as cross-covariances across different climate models. Our method allows for a nonseparable and nonstationary cross-covariance structure. We also present a covariance approximation approach to facilitate the computation in the modeling and analysis of very large multivariate spatial data sets. The covariance approximation consists of two parts: a reduced-rank part to capture the large-scale spatial dependence, and a sparse covariance matrix to correct the small-scale dependence error induced by the reduced rank approximation. We pay special attention to the case that the second part of the approximation has a block-diagonal structure. Simulation results of model fitting and prediction show substantial improvement of the proposed approximation over the predictive process approximation and the independent blocks analysis. We then apply our computational approach to the joint statistical modeling of multiple climate model errors.
Full work available at URL: https://arxiv.org/abs/1203.0133
Recommendations
- A Full Scale Approximation of Covariance Functions for Large Spatial Data Sets
- Cross-covariance functions for multivariate geostatistics
- Non-stationary cross-covariance models for multivariate processes on a globe
- Covariance Modeling for Multivariate Spatial Processes Based on Separable Approximations
- Matérn-based nonstationary cross-covariance models for global processes
Bayesian inference (62F15) Directional data; spatial statistics (62H11) Inference from spatial processes (62M30) Applications of statistics to environmental and related topics (62P12)
Cites Work
- Gaussian Markov Random Fields
- Fixed Rank Kriging for Very Large Spatial Data Sets
- Finding Groups in Data
- Bayesian Measures of Model Complexity and Fit
- A modeling approach for large spatial datasets
- Compactly supported correlation functions
- Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree
- Gaussian Predictive Process Models for Large Spatial Data Sets
- Fitting Gaussian Markov Random Fields to Gaussian Fields
- Title not available (Why is that?)
- Approximating Likelihoods for Large Spatial Data Sets
- Title not available (Why is that?)
- Matérn Cross-Covariance Functions for Multivariate Random Fields
- Approximate Likelihood for Large Irregularly Spaced Spatial Data
- Covariance Tapering for Likelihood-Based Estimation in Large Spatial Data Sets
- A spatial analysis of multivariate output from regional climate models
- Title not available (Why is that?)
- Sampling-Based Approaches to Calculating Marginal Densities
- Title not available (Why is that?)
- Improving the performance of predictive process modeling for large datasets
- Title not available (Why is that?)
- Nonstationary multivariate process modeling through spatially varying coregionalization
- Multivariate spatial modeling for geostatistical data using convolved covariance functions
- Title not available (Why is that?)
- Cross-covariance functions for multivariate random fields based on latent dimensions
- Error estimates for interpolation by compactly supported radial basis functions of minimal degree
- A dimension-reduced approach to space-time Kalman filtering
- Hierarchical Spatial Process Models for Multiple Traits in Large Genetic Trials
- Space and space-time modeling using process convolutions
- Case studies in Bayesian statistics. Vol 6: Proceedings of the 6th workshop, Carnegie Mellon University, Pittsburgh, PA, USA, September 28--29, 2001
- Computing Dirichlet Tessellations in the Plane
- Bayesian Modeling of Uncertainty in Ensembles of Climate Models
- Spatial analysis to quantify numerical model bias and dependence
- Geoadditive Models
- Latent variable modeling for integrating output from multiple climate models
- Asymptotic properties of computationally efficient alternative estimators for a class of multivariate normal models
Cited In (27)
- Bayesian nonstationary spatial modeling for very large datasets
- Vecchia-Laplace approximations of generalized Gaussian processes for big non-Gaussian spatial data
- Multi-scale Vecchia approximations of Gaussian processes
- Spatial Linear Regression with Covariate Measurement Errors: Inference and Scalable Computation in a Functional Modeling Approach
- A case study competition among methods for analyzing large spatial data
- Bayesian modeling of discrete-time point-referenced spatio-temporal data
- A Bayesian spatial factor analysis approach for combining climate model ensembles
- Matérn-based nonstationary cross-covariance models for global processes
- Regularity theory for a new class of fractional parabolic stochastic evolution equations
- Scalable penalized spatiotemporal land-use regression for ground-level nitrogen dioxide
- Scalable computations for nonstationary Gaussian processes
- Bayesian fixed-domain asymptotics for covariance parameters in a Gaussian process model
- Full scale multi-output Gaussian process emulator with nonseparable auto-covariance functions
- A large-scale spatio-temporal binomial regression model for estimating seroprevalence trends
- Conjugate sparse plus low rank models for efficient Bayesian interpolation of large spatial data
- Time-varying correlation structure estimation and local-feature detection for spatio-temporal data
- Efficient Algorithms for Bayesian Nearest Neighbor Gaussian Processes
- A general framework for Vecchia approximations of Gaussian processes
- Rejoinder
- Modeling complex spatial dependencies: low-rank spatially varying cross-covariances with application to soil nutrient data
- Spatial regression with non-parametric modeling of Fourier coefficients
- Correlation-based sparse inverse Cholesky factorization for fast Gaussian-process inference
- Parallel inference for massive distributed spatial data using low-rank models
- Cross-covariance functions for multivariate geostatistics
- Predictive and sensitive analysis of a bivariate skewed spatial process based on the Bayesian framework
- 30 years of space-time covariance functions
- Multivariate modelling of spatial extremes based on copulas
Uses Software
This page was built for publication: Covariance approximation for large multivariate spatial data sets with an application to multiple climate model errors
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q765994)
