Modified linear projection for large spatial datasets

DOI10.1080/03610918.2014.983649MaRDI QIDQ2974895zbMATH OpenOpenAlexFDO

Authors Toshihiro Hirano

Publication date 11 April 2017

Published in Communications in Statistics. Simulation and Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1402.5847

zbMATH Keywords

covariance taperingGaussian processgeostatisticsMarkov chain Monte Carloreduced rank approximationstochastic matrix approximation

Mathematics Subject Classification ID

Bayesian inference (62F15) Inference from spatial processes (62M30)

Abstract: Recent developments in engineering techniques for spatial data collection such as geographic information systems have resulted in an increasing need for methods to analyze large spatial data sets. These sorts of data sets can be found in various fields of the natural and social sciences. However, model fitting and spatial prediction using these large spatial data sets are impractically time-consuming, because of the necessary matrix inversions. Various methods have been developed to deal with this problem, including a reduced rank approach and a sparse matrix approximation. In this paper, we propose a modification to an existing reduced rank approach to capture both the large- and small-scale spatial variations effectively. We have used simulated examples and an empirical data analysis to demonstrate that our proposed approach consistently performs well when compared with other methods. In particular, the performance of our new method does not depend on the dependence properties of the spatial covariance functions.

Recommendations

• A multi-resolution approximation via linear projection for large spatial datasets
• Improving the performance of predictive process modeling for large datasets
• A case study competition among methods for analyzing large spatial data
• Large spatial data modeling and analysis: a Krylov subspace approach
• Statistical modeling of spatial big data: an approach from a functional data analysis perspective

Cited in

(5)

• Statistical modeling of spatial big data: an approach from a functional data analysis perspective
• Spatial data compression via adaptive dispersion clustering
• A case study competition among methods for analyzing large spatial data
• A multi-resolution approximation via linear projection for large spatial datasets
• Improving the performance of predictive process modeling for large datasets

This page was built for publication: Modified 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 Q2974895)