A tuning parameter free test for properties of space-time covariance functions

Research output: Contribution to journalArticle

Abstract

We propose a new nonparametric test to test for symmetry and separability of space-time covariance functions. Unlike the existing nonparametric tests, our test has the attractive convenience of being free of choosing any user-chosen number or smoothing parameter. The asymptotic null distributions of the test statistics are free of nuisance parameters and the critical values have been tabulated in the literature. From a practical point of view, our test is easy to implement and can be readily used by the practitioner. A Monte-Carlo experiment and real data analysis illustrate the finite sample performance of the new test.

Original languageEnglish (US)
Pages (from-to)4031-4038
Number of pages8
JournalJournal of Statistical Planning and Inference
Volume139
Issue number12
DOIs
StatePublished - Dec 1 2009

Fingerprint

Covariance Function
Parameter Tuning
Tuning
Space-time
Statistics
Non-parametric test
Experiments
Smoothing Parameter
Monte Carlo Experiment
Nuisance Parameter
Null Distribution
Separability
Asymptotic distribution
Test Statistic
Critical value
Data analysis
Symmetry
Nonparametric test

Keywords

  • Asymptotically pivotal
  • Covariance
  • Full symmetry
  • Separability

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

A tuning parameter free test for properties of space-time covariance functions. / Shao, Xiaofeng; Li, Bo.

In: Journal of Statistical Planning and Inference, Vol. 139, No. 12, 01.12.2009, p. 4031-4038.

Research output: Contribution to journalArticle

@article{c725847089b54a72a95aea340085a961,
title = "A tuning parameter free test for properties of space-time covariance functions",
abstract = "We propose a new nonparametric test to test for symmetry and separability of space-time covariance functions. Unlike the existing nonparametric tests, our test has the attractive convenience of being free of choosing any user-chosen number or smoothing parameter. The asymptotic null distributions of the test statistics are free of nuisance parameters and the critical values have been tabulated in the literature. From a practical point of view, our test is easy to implement and can be readily used by the practitioner. A Monte-Carlo experiment and real data analysis illustrate the finite sample performance of the new test.",
keywords = "Asymptotically pivotal, Covariance, Full symmetry, Separability",
author = "Xiaofeng Shao and Bo Li",
year = "2009",
month = "12",
day = "1",
doi = "10.1016/j.jspi.2009.05.008",
language = "English (US)",
volume = "139",
pages = "4031--4038",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier",
number = "12",

}

TY - JOUR

T1 - A tuning parameter free test for properties of space-time covariance functions

AU - Shao, Xiaofeng

AU - Li, Bo

PY - 2009/12/1

Y1 - 2009/12/1

N2 - We propose a new nonparametric test to test for symmetry and separability of space-time covariance functions. Unlike the existing nonparametric tests, our test has the attractive convenience of being free of choosing any user-chosen number or smoothing parameter. The asymptotic null distributions of the test statistics are free of nuisance parameters and the critical values have been tabulated in the literature. From a practical point of view, our test is easy to implement and can be readily used by the practitioner. A Monte-Carlo experiment and real data analysis illustrate the finite sample performance of the new test.

AB - We propose a new nonparametric test to test for symmetry and separability of space-time covariance functions. Unlike the existing nonparametric tests, our test has the attractive convenience of being free of choosing any user-chosen number or smoothing parameter. The asymptotic null distributions of the test statistics are free of nuisance parameters and the critical values have been tabulated in the literature. From a practical point of view, our test is easy to implement and can be readily used by the practitioner. A Monte-Carlo experiment and real data analysis illustrate the finite sample performance of the new test.

KW - Asymptotically pivotal

KW - Covariance

KW - Full symmetry

KW - Separability

UR - http://www.scopus.com/inward/record.url?scp=69249222573&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69249222573&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2009.05.008

DO - 10.1016/j.jspi.2009.05.008

M3 - Article

VL - 139

SP - 4031

EP - 4038

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

IS - 12

ER -