An approach to modeling asymmetric multivariate spatial covariance structures

Bo Li, Hao Zhang

Research output: Contribution to journalArticle

Abstract

We propose a framework in light of the delay effect to model the asymmetry of multivariate covariance functions that is often exhibited in real data. This general approach can endow any valid symmetric multivariate covariance function with the ability of modeling asymmetry and is very easy to implement. Our simulations and real data examples show that asymmetric multivariate covariance functions based on our approach can achieve remarkable improvements in prediction over symmetric models.

Original languageEnglish (US)
Pages (from-to)1445-1453
Number of pages9
JournalJournal of Multivariate Analysis
Volume102
Issue number10
DOIs
StatePublished - Nov 1 2011

Fingerprint

Multivariate Functions
Covariance Function
Covariance Structure
Spatial Structure
Asymmetry
Modeling
Valid
Prediction
Model
Simulation

Keywords

  • Asymmetry
  • Bivariate matérn
  • Intrinsic model
  • Multivariate covariance function
  • Symmetry

ASJC Scopus subject areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

Cite this

An approach to modeling asymmetric multivariate spatial covariance structures. / Li, Bo; Zhang, Hao.

In: Journal of Multivariate Analysis, Vol. 102, No. 10, 01.11.2011, p. 1445-1453.

Research output: Contribution to journalArticle

@article{625dd571cfed42aea5fb73cf99f7ecbf,
title = "An approach to modeling asymmetric multivariate spatial covariance structures",
abstract = "We propose a framework in light of the delay effect to model the asymmetry of multivariate covariance functions that is often exhibited in real data. This general approach can endow any valid symmetric multivariate covariance function with the ability of modeling asymmetry and is very easy to implement. Our simulations and real data examples show that asymmetric multivariate covariance functions based on our approach can achieve remarkable improvements in prediction over symmetric models.",
keywords = "Asymmetry, Bivariate mat{\'e}rn, Intrinsic model, Multivariate covariance function, Symmetry",
author = "Bo Li and Hao Zhang",
year = "2011",
month = "11",
day = "1",
doi = "10.1016/j.jmva.2011.05.010",
language = "English (US)",
volume = "102",
pages = "1445--1453",
journal = "Journal of Multivariate Analysis",
issn = "0047-259X",
publisher = "Academic Press Inc.",
number = "10",

}

TY - JOUR

T1 - An approach to modeling asymmetric multivariate spatial covariance structures

AU - Li, Bo

AU - Zhang, Hao

PY - 2011/11/1

Y1 - 2011/11/1

N2 - We propose a framework in light of the delay effect to model the asymmetry of multivariate covariance functions that is often exhibited in real data. This general approach can endow any valid symmetric multivariate covariance function with the ability of modeling asymmetry and is very easy to implement. Our simulations and real data examples show that asymmetric multivariate covariance functions based on our approach can achieve remarkable improvements in prediction over symmetric models.

AB - We propose a framework in light of the delay effect to model the asymmetry of multivariate covariance functions that is often exhibited in real data. This general approach can endow any valid symmetric multivariate covariance function with the ability of modeling asymmetry and is very easy to implement. Our simulations and real data examples show that asymmetric multivariate covariance functions based on our approach can achieve remarkable improvements in prediction over symmetric models.

KW - Asymmetry

KW - Bivariate matérn

KW - Intrinsic model

KW - Multivariate covariance function

KW - Symmetry

UR - http://www.scopus.com/inward/record.url?scp=79960050538&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960050538&partnerID=8YFLogxK

U2 - 10.1016/j.jmva.2011.05.010

DO - 10.1016/j.jmva.2011.05.010

M3 - Article

VL - 102

SP - 1445

EP - 1453

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

SN - 0047-259X

IS - 10

ER -