Self-normalization for Spatial Data

Xianyang Zhang, Bo Li, Xiaofeng Shao

Research output: Contribution to journalArticle

Abstract

This paper considers inference for both spatial lattice data with possibly irregularly shaped sampling region and non-lattice data, by extending the recently proposed self-normalization (SN) approach from stationary time series to the spatial setup. A nice feature of the SN method is that it avoids the choice of tuning parameters, which are usually required for other non-parametric inference approaches. The extension is non-trivial as spatial data has no natural one-directional time ordering. The SN-based inference is convenient to implement and is shown through simulation studies to provide more accurate coverage compared with the widely used subsampling approach. We also illustrate the idea of SN using a real data example.

Original languageEnglish (US)
Pages (from-to)311-324
Number of pages14
JournalScandinavian Journal of Statistics
Volume41
Issue number2
DOIs
StatePublished - Jan 1 2014

Fingerprint

Self-normalization
Spatial Data
Nonparametric Inference
Stationary Time Series
Subsampling
Parameter Tuning
Coverage
Simulation Study
Normalization
Inference

Keywords

  • Increasing domain asymptotics
  • Random field
  • Self-normalization
  • Spatial point process
  • Subsampling

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Self-normalization for Spatial Data. / Zhang, Xianyang; Li, Bo; Shao, Xiaofeng.

In: Scandinavian Journal of Statistics, Vol. 41, No. 2, 01.01.2014, p. 311-324.

Research output: Contribution to journalArticle

@article{ae0c488fb8ad4e6ab4d6295154f9a93d,
title = "Self-normalization for Spatial Data",
abstract = "This paper considers inference for both spatial lattice data with possibly irregularly shaped sampling region and non-lattice data, by extending the recently proposed self-normalization (SN) approach from stationary time series to the spatial setup. A nice feature of the SN method is that it avoids the choice of tuning parameters, which are usually required for other non-parametric inference approaches. The extension is non-trivial as spatial data has no natural one-directional time ordering. The SN-based inference is convenient to implement and is shown through simulation studies to provide more accurate coverage compared with the widely used subsampling approach. We also illustrate the idea of SN using a real data example.",
keywords = "Increasing domain asymptotics, Random field, Self-normalization, Spatial point process, Subsampling",
author = "Xianyang Zhang and Bo Li and Xiaofeng Shao",
year = "2014",
month = "1",
day = "1",
doi = "10.1111/sjos.12028",
language = "English (US)",
volume = "41",
pages = "311--324",
journal = "Scandinavian Journal of Statistics",
issn = "0303-6898",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Self-normalization for Spatial Data

AU - Zhang, Xianyang

AU - Li, Bo

AU - Shao, Xiaofeng

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper considers inference for both spatial lattice data with possibly irregularly shaped sampling region and non-lattice data, by extending the recently proposed self-normalization (SN) approach from stationary time series to the spatial setup. A nice feature of the SN method is that it avoids the choice of tuning parameters, which are usually required for other non-parametric inference approaches. The extension is non-trivial as spatial data has no natural one-directional time ordering. The SN-based inference is convenient to implement and is shown through simulation studies to provide more accurate coverage compared with the widely used subsampling approach. We also illustrate the idea of SN using a real data example.

AB - This paper considers inference for both spatial lattice data with possibly irregularly shaped sampling region and non-lattice data, by extending the recently proposed self-normalization (SN) approach from stationary time series to the spatial setup. A nice feature of the SN method is that it avoids the choice of tuning parameters, which are usually required for other non-parametric inference approaches. The extension is non-trivial as spatial data has no natural one-directional time ordering. The SN-based inference is convenient to implement and is shown through simulation studies to provide more accurate coverage compared with the widely used subsampling approach. We also illustrate the idea of SN using a real data example.

KW - Increasing domain asymptotics

KW - Random field

KW - Self-normalization

KW - Spatial point process

KW - Subsampling

UR - http://www.scopus.com/inward/record.url?scp=84899951932&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899951932&partnerID=8YFLogxK

U2 - 10.1111/sjos.12028

DO - 10.1111/sjos.12028

M3 - Article

VL - 41

SP - 311

EP - 324

JO - Scandinavian Journal of Statistics

JF - Scandinavian Journal of Statistics

SN - 0303-6898

IS - 2

ER -