Beyond the valley of the covariance function

Daniel Simpson, Finn Lindgren, Håvard Rue

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Multivariate models are under-represented in the literature on spatial statistics. There is a basic reason for this: univariate models are sufficiently complicated to
keep us busy. Genton and Kleiber have done a fabulous job compiling and investigating the available models, with a focus on the important class of models that they, with collaborators, introduced. This paper gives a solid state of the art and points out just how many holes there are in the theory and practice associated with these fields. This gives us licence to point out some other holes and to suggest some important directions for the future.
Original languageEnglish
Pages (from-to)164-166
Number of pages3
JournalStatistical Science
Issue number2
Publication statusPublished - 30 Jul 2015

Keywords / Materials (for Non-textual outputs)

  • stat.ME


Dive into the research topics of 'Beyond the valley of the covariance function'. Together they form a unique fingerprint.

Cite this