On the effect of overdispersion on exact conditional tests

Thomas A. Severini

doi:10.1111/1467-9868.00166

On the effect of overdispersion on exact conditional tests

Thomas A. Severini^*

^*Corresponding author for this work

Statistics and Data Science

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Let Y₁, . . ., Y_n denote independent random variables such that Y_j has a one-parameter exponential family distribution with canonical parameter θ_j = λ + ψX_j, here X₁, . . ., X_n are known constants. Consider a test of the null hypothesis ψ = 0. Under the null hypothesis, A = Σ Y_j is sufficient for λ and, hence, a test of ψ = 0 may be based on the conditional distribution of T = Σ X_jY_j given A, which is independent of λ. In this paper, the effects of overdispersion due to a mixture model on the conditional distribution of T given A are considered.

Original language	English (US)
Pages (from-to)	115-126
Number of pages	12
Journal	Journal of the Royal Statistical Society. Series B: Statistical Methodology
Volume	61
Issue number	1
DOIs	https://doi.org/10.1111/1467-9868.00166
State	Published - 1999

Keywords

Asymptotic expansions
Conditional inference
Exact tests
Mixture models
Robustness
Significance tests

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1111/1467-9868.00166

Cite this

@article{6a931021cd8d4e6e9fcc23cfe098b410,

title = "On the effect of overdispersion on exact conditional tests",

abstract = "Let Y1, . . ., Yn denote independent random variables such that Yj has a one-parameter exponential family distribution with canonical parameter θj = λ + ψXj, here X1, . . ., Xn are known constants. Consider a test of the null hypothesis ψ = 0. Under the null hypothesis, A = Σ Yj is sufficient for λ and, hence, a test of ψ = 0 may be based on the conditional distribution of T = Σ XjYj given A, which is independent of λ. In this paper, the effects of overdispersion due to a mixture model on the conditional distribution of T given A are considered.",

keywords = "Asymptotic expansions, Conditional inference, Exact tests, Mixture models, Robustness, Significance tests",

author = "Severini, {Thomas A.}",

year = "1999",

doi = "10.1111/1467-9868.00166",

language = "English (US)",

volume = "61",

pages = "115--126",

journal = "Journal of the Royal Statistical Society. Series B: Statistical Methodology",

issn = "1369-7412",

publisher = "Oxford University Press",

number = "1",

}

TY - JOUR

T1 - On the effect of overdispersion on exact conditional tests

AU - Severini, Thomas A.

PY - 1999

Y1 - 1999

N2 - Let Y1, . . ., Yn denote independent random variables such that Yj has a one-parameter exponential family distribution with canonical parameter θj = λ + ψXj, here X1, . . ., Xn are known constants. Consider a test of the null hypothesis ψ = 0. Under the null hypothesis, A = Σ Yj is sufficient for λ and, hence, a test of ψ = 0 may be based on the conditional distribution of T = Σ XjYj given A, which is independent of λ. In this paper, the effects of overdispersion due to a mixture model on the conditional distribution of T given A are considered.

AB - Let Y1, . . ., Yn denote independent random variables such that Yj has a one-parameter exponential family distribution with canonical parameter θj = λ + ψXj, here X1, . . ., Xn are known constants. Consider a test of the null hypothesis ψ = 0. Under the null hypothesis, A = Σ Yj is sufficient for λ and, hence, a test of ψ = 0 may be based on the conditional distribution of T = Σ XjYj given A, which is independent of λ. In this paper, the effects of overdispersion due to a mixture model on the conditional distribution of T given A are considered.

KW - Asymptotic expansions

KW - Conditional inference

KW - Exact tests

KW - Mixture models

KW - Robustness

KW - Significance tests

UR - http://www.scopus.com/inward/record.url?scp=0033473460&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033473460&partnerID=8YFLogxK

U2 - 10.1111/1467-9868.00166

DO - 10.1111/1467-9868.00166

M3 - Article

AN - SCOPUS:0033473460

SN - 1369-7412

VL - 61

SP - 115

EP - 126

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

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

IS - 1

ER -

On the effect of overdispersion on exact conditional tests

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this