Detection of change in the spatiotemporal mean function

Oleksandr Gromenko, Piotr Kokoszka, Matthew Logan Reimherr

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The paper develops inferential methodology for detecting a change in the annual pattern of an environmental variable measured at fixed locations in a spatial region. Using a framework built on functional data analysis, we model observations as a collection of function-valued time sequences available at many sites. Each sequence is modelled as an annual mean function, which may change, plus a sequence of error functions, which are spatially correlated. The tests statistics extend the cumulative sum paradigm to this more complex setting. Their asymptotic distributions are not parameter free because of the spatial dependence but can be effectively approximated by Monte Carlo simulations. The new methodology is applied to precipitation data. Its finite sample performance is assessed by a simulation study.

Original languageEnglish (US)
Pages (from-to)29-50
Number of pages22
JournalJournal of the Royal Statistical Society. Series B: Statistical Methodology
Volume79
Issue number1
DOIs
StatePublished - Jan 1 2017

Fingerprint

Annual
Functional Data Analysis
Spatial Dependence
Cumulative Sum
Methodology
Error function
Asymptotic distribution
Test Statistic
Monte Carlo Simulation
Paradigm
Simulation Study
Model
Framework
Observation
Monte Carlo simulation
Finite sample
Simulation study
Cumulative sum
Test statistic
Spatial dependence

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{ec3143c3dc3d4cb2af12f24c8b289700,
title = "Detection of change in the spatiotemporal mean function",
abstract = "The paper develops inferential methodology for detecting a change in the annual pattern of an environmental variable measured at fixed locations in a spatial region. Using a framework built on functional data analysis, we model observations as a collection of function-valued time sequences available at many sites. Each sequence is modelled as an annual mean function, which may change, plus a sequence of error functions, which are spatially correlated. The tests statistics extend the cumulative sum paradigm to this more complex setting. Their asymptotic distributions are not parameter free because of the spatial dependence but can be effectively approximated by Monte Carlo simulations. The new methodology is applied to precipitation data. Its finite sample performance is assessed by a simulation study.",
author = "Oleksandr Gromenko and Piotr Kokoszka and Reimherr, {Matthew Logan}",
year = "2017",
month = "1",
day = "1",
doi = "10.1111/rssb.12156",
language = "English (US)",
volume = "79",
pages = "29--50",
journal = "Journal of the Royal Statistical Society. Series B: Statistical Methodology",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "1",

}

Detection of change in the spatiotemporal mean function. / Gromenko, Oleksandr; Kokoszka, Piotr; Reimherr, Matthew Logan.

In: Journal of the Royal Statistical Society. Series B: Statistical Methodology, Vol. 79, No. 1, 01.01.2017, p. 29-50.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Detection of change in the spatiotemporal mean function

AU - Gromenko, Oleksandr

AU - Kokoszka, Piotr

AU - Reimherr, Matthew Logan

PY - 2017/1/1

Y1 - 2017/1/1

N2 - The paper develops inferential methodology for detecting a change in the annual pattern of an environmental variable measured at fixed locations in a spatial region. Using a framework built on functional data analysis, we model observations as a collection of function-valued time sequences available at many sites. Each sequence is modelled as an annual mean function, which may change, plus a sequence of error functions, which are spatially correlated. The tests statistics extend the cumulative sum paradigm to this more complex setting. Their asymptotic distributions are not parameter free because of the spatial dependence but can be effectively approximated by Monte Carlo simulations. The new methodology is applied to precipitation data. Its finite sample performance is assessed by a simulation study.

AB - The paper develops inferential methodology for detecting a change in the annual pattern of an environmental variable measured at fixed locations in a spatial region. Using a framework built on functional data analysis, we model observations as a collection of function-valued time sequences available at many sites. Each sequence is modelled as an annual mean function, which may change, plus a sequence of error functions, which are spatially correlated. The tests statistics extend the cumulative sum paradigm to this more complex setting. Their asymptotic distributions are not parameter free because of the spatial dependence but can be effectively approximated by Monte Carlo simulations. The new methodology is applied to precipitation data. Its finite sample performance is assessed by a simulation study.

UR - http://www.scopus.com/inward/record.url?scp=84959421804&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959421804&partnerID=8YFLogxK

U2 - 10.1111/rssb.12156

DO - 10.1111/rssb.12156

M3 - Article

AN - SCOPUS:84959421804

VL - 79

SP - 29

EP - 50

JO - Journal of the Royal Statistical Society. Series B: Statistical Methodology

JF - Journal of the Royal Statistical Society. Series B: Statistical Methodology

SN - 1369-7412

IS - 1

ER -