Standard Article

Massive Data, Models for

Spatial and Temporal Modeling and Analysis

  1. Hao Zhang

Published Online: 15 JAN 2013

DOI: 10.1002/9780470057339.vnn053

Encyclopedia of Environmetrics

Encyclopedia of Environmetrics

How to Cite

Zhang, H. 2013. Massive Data, Models for. Encyclopedia of Environmetrics. 3.

Author Information

  1. Purdue University, West Lafayette, IN, USA

Publication History

  1. Published Online: 15 JAN 2013

Abstract

Massive data present a challenge to the use and manipulation of well-studied inferential strategies such as maximum likelihood estimation, Bayesian methods, and best linear prediction (kriging) since these methods require the inversion of a large covariance matrix. This entry introduces statistical models that circumvent computational issues, which include local models, covariance tapering, low-rank models, and Markovian models.

Keywords:

  • covariance tapering;
  • covariogram;
  • kriging;
  • low-rank model;
  • massive data;
  • sparse matrix