Modeling nonstationarity in space and time

Lyndsay Shand, Bo Li

Research output: Contribution to journalArticle

Abstract

We propose to model a spatio-temporal random field that has nonstationary covariance structure in both space and time domains by applying the concept of the dimension expansion method in Bornn et al. (2012). Simulations are conducted for both separable and nonseparable space-time covariance models, and the model is also illustrated with a streamflow dataset. Both simulation and data analyses show that modeling nonstationarity in both space and time can improve the predictive performance over stationary covariance models or models that are nonstationary in space but stationary in time.

Original languageEnglish (US)
Pages (from-to)759-768
Number of pages10
JournalBiometrics
Volume73
Issue number3
DOIs
StatePublished - Sep 1 2017

Fingerprint

Nonstationarity
space and time
Modeling
Model
Nonseparable
Covariance Structure
Random Field
stream flow
Time Domain
Simulation
Space-time

Keywords

  • Dimension expansion
  • Nonstationarity
  • Space-time random field

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

Modeling nonstationarity in space and time. / Shand, Lyndsay; Li, Bo.

In: Biometrics, Vol. 73, No. 3, 01.09.2017, p. 759-768.

Research output: Contribution to journalArticle

Shand, Lyndsay ; Li, Bo. / Modeling nonstationarity in space and time. In: Biometrics. 2017 ; Vol. 73, No. 3. pp. 759-768.
@article{21779a8b33994805af5f312d9d63c6e5,
title = "Modeling nonstationarity in space and time",
abstract = "We propose to model a spatio-temporal random field that has nonstationary covariance structure in both space and time domains by applying the concept of the dimension expansion method in Bornn et al. (2012). Simulations are conducted for both separable and nonseparable space-time covariance models, and the model is also illustrated with a streamflow dataset. Both simulation and data analyses show that modeling nonstationarity in both space and time can improve the predictive performance over stationary covariance models or models that are nonstationary in space but stationary in time.",
keywords = "Dimension expansion, Nonstationarity, Space-time random field",
author = "Lyndsay Shand and Bo Li",
year = "2017",
month = "9",
day = "1",
doi = "10.1111/biom.12656",
language = "English (US)",
volume = "73",
pages = "759--768",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Modeling nonstationarity in space and time

AU - Shand, Lyndsay

AU - Li, Bo

PY - 2017/9/1

Y1 - 2017/9/1

N2 - We propose to model a spatio-temporal random field that has nonstationary covariance structure in both space and time domains by applying the concept of the dimension expansion method in Bornn et al. (2012). Simulations are conducted for both separable and nonseparable space-time covariance models, and the model is also illustrated with a streamflow dataset. Both simulation and data analyses show that modeling nonstationarity in both space and time can improve the predictive performance over stationary covariance models or models that are nonstationary in space but stationary in time.

AB - We propose to model a spatio-temporal random field that has nonstationary covariance structure in both space and time domains by applying the concept of the dimension expansion method in Bornn et al. (2012). Simulations are conducted for both separable and nonseparable space-time covariance models, and the model is also illustrated with a streamflow dataset. Both simulation and data analyses show that modeling nonstationarity in both space and time can improve the predictive performance over stationary covariance models or models that are nonstationary in space but stationary in time.

KW - Dimension expansion

KW - Nonstationarity

KW - Space-time random field

UR - http://www.scopus.com/inward/record.url?scp=85011573520&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011573520&partnerID=8YFLogxK

U2 - 10.1111/biom.12656

DO - 10.1111/biom.12656

M3 - Article

VL - 73

SP - 759

EP - 768

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 3

ER -