Testing for positive expectation dependence

Xuehu Zhu, Xu Guo, Lu Lin, Lixing Zhu

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In this paper, hypothesis testing for positive first-degree and higher-degree expectation dependence is investigated. Some tests of Kolmogorov–Smirnov type are constructed, which are shown to control type I error well and to be consistent against global alternative hypothesis. Further, the tests can also detect local alternative hypotheses distinct from the null hypothesis at a rate as close to the square root of the sample size as possible, which is the fastest possible rate in hypothesis testing. A nonparametric Monte Carlo test procedure is applied to implement the new tests because both sampling and limiting null distributions are not tractable. Simulation studies and a real data analysis are carried out to illustrate the performances of the new tests.

Original languageEnglish (US)
Pages (from-to)135-153
Number of pages19
JournalAnnals of the Institute of Statistical Mathematics
Volume68
Issue number1
DOIs
StatePublished - Feb 1 2016

Fingerprint

Testing
Hypothesis Testing
Monte Carlo Test
Local Alternatives
Type I error
Non-parametric test
Null Distribution
Limiting Distribution
Square root
Null hypothesis
Data analysis
Sample Size
Simulation Study
Distinct
Alternatives

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

Zhu, Xuehu ; Guo, Xu ; Lin, Lu ; Zhu, Lixing. / Testing for positive expectation dependence. In: Annals of the Institute of Statistical Mathematics. 2016 ; Vol. 68, No. 1. pp. 135-153.
@article{63db50e5d51340778fed61a5ffd9f119,
title = "Testing for positive expectation dependence",
abstract = "In this paper, hypothesis testing for positive first-degree and higher-degree expectation dependence is investigated. Some tests of Kolmogorov–Smirnov type are constructed, which are shown to control type I error well and to be consistent against global alternative hypothesis. Further, the tests can also detect local alternative hypotheses distinct from the null hypothesis at a rate as close to the square root of the sample size as possible, which is the fastest possible rate in hypothesis testing. A nonparametric Monte Carlo test procedure is applied to implement the new tests because both sampling and limiting null distributions are not tractable. Simulation studies and a real data analysis are carried out to illustrate the performances of the new tests.",
author = "Xuehu Zhu and Xu Guo and Lu Lin and Lixing Zhu",
year = "2016",
month = "2",
day = "1",
doi = "10.1007/s10463-014-0492-7",
language = "English (US)",
volume = "68",
pages = "135--153",
journal = "Annals of the Institute of Statistical Mathematics",
issn = "0020-3157",
publisher = "Springer Netherlands",
number = "1",

}

Testing for positive expectation dependence. / Zhu, Xuehu; Guo, Xu; Lin, Lu; Zhu, Lixing.

In: Annals of the Institute of Statistical Mathematics, Vol. 68, No. 1, 01.02.2016, p. 135-153.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Testing for positive expectation dependence

AU - Zhu, Xuehu

AU - Guo, Xu

AU - Lin, Lu

AU - Zhu, Lixing

PY - 2016/2/1

Y1 - 2016/2/1

N2 - In this paper, hypothesis testing for positive first-degree and higher-degree expectation dependence is investigated. Some tests of Kolmogorov–Smirnov type are constructed, which are shown to control type I error well and to be consistent against global alternative hypothesis. Further, the tests can also detect local alternative hypotheses distinct from the null hypothesis at a rate as close to the square root of the sample size as possible, which is the fastest possible rate in hypothesis testing. A nonparametric Monte Carlo test procedure is applied to implement the new tests because both sampling and limiting null distributions are not tractable. Simulation studies and a real data analysis are carried out to illustrate the performances of the new tests.

AB - In this paper, hypothesis testing for positive first-degree and higher-degree expectation dependence is investigated. Some tests of Kolmogorov–Smirnov type are constructed, which are shown to control type I error well and to be consistent against global alternative hypothesis. Further, the tests can also detect local alternative hypotheses distinct from the null hypothesis at a rate as close to the square root of the sample size as possible, which is the fastest possible rate in hypothesis testing. A nonparametric Monte Carlo test procedure is applied to implement the new tests because both sampling and limiting null distributions are not tractable. Simulation studies and a real data analysis are carried out to illustrate the performances of the new tests.

UR - http://www.scopus.com/inward/record.url?scp=84953358440&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84953358440&partnerID=8YFLogxK

U2 - 10.1007/s10463-014-0492-7

DO - 10.1007/s10463-014-0492-7

M3 - Article

AN - SCOPUS:84953358440

VL - 68

SP - 135

EP - 153

JO - Annals of the Institute of Statistical Mathematics

JF - Annals of the Institute of Statistical Mathematics

SN - 0020-3157

IS - 1

ER -