An optimal property of the exact multinomial test and the extended Fisher's exact test

Seung Ho Kang

doi:10.1016/S0167-7152(99)00010-3

An optimal property of the exact multinomial test and the extended Fisher's exact test

Seung Ho Kang

Department of Applied Statistics

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

2 Citations (Scopus)

Abstract

In the goodness-of-fit test of parameters of the multinomial distribution we show that the exact multinomial test is asymptotically equivalent to the likelihood ratio test by using Stirling's formula. In an r×c contingency table, we show that the extended Fisher's exact test conditional on row and column margins for the test of independence is also asymptotically equivalent to the likelihood ratio test. From the Bahadur asymptotic optimality of the likelihood ratio test in both unconditional and conditional cases, we prove that the two exact tests are asymptotically optimal in the sense of Bahadur efficiency.

Original language	English
Pages (from-to)	201-207
Number of pages	7
Journal	Statistics and Probability Letters
Volume	44
Issue number	2
DOIs	https://doi.org/10.1016/S0167-7152(99)00010-3
Publication status	Published - 1999 Aug 15

Bibliographical note

Funding Information:
This research was supported in part by an appointment to the Postgraduate Research Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration.

All Science Journal Classification (ASJC) codes

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1016/S0167-7152(99)00010-3

Cite this

@article{c5eaf03ee1e24d8187e12a91815b5b90,

title = "An optimal property of the exact multinomial test and the extended Fisher's exact test",

abstract = "In the goodness-of-fit test of parameters of the multinomial distribution we show that the exact multinomial test is asymptotically equivalent to the likelihood ratio test by using Stirling's formula. In an r×c contingency table, we show that the extended Fisher's exact test conditional on row and column margins for the test of independence is also asymptotically equivalent to the likelihood ratio test. From the Bahadur asymptotic optimality of the likelihood ratio test in both unconditional and conditional cases, we prove that the two exact tests are asymptotically optimal in the sense of Bahadur efficiency.",

author = "Kang, {Seung Ho}",

note = "Funding Information: This research was supported in part by an appointment to the Postgraduate Research Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration. ",

year = "1999",

month = aug,

day = "15",

doi = "10.1016/S0167-7152(99)00010-3",

language = "English",

volume = "44",

pages = "201--207",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - An optimal property of the exact multinomial test and the extended Fisher's exact test

AU - Kang, Seung Ho

N1 - Funding Information: This research was supported in part by an appointment to the Postgraduate Research Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug Administration.

PY - 1999/8/15

Y1 - 1999/8/15

N2 - In the goodness-of-fit test of parameters of the multinomial distribution we show that the exact multinomial test is asymptotically equivalent to the likelihood ratio test by using Stirling's formula. In an r×c contingency table, we show that the extended Fisher's exact test conditional on row and column margins for the test of independence is also asymptotically equivalent to the likelihood ratio test. From the Bahadur asymptotic optimality of the likelihood ratio test in both unconditional and conditional cases, we prove that the two exact tests are asymptotically optimal in the sense of Bahadur efficiency.

AB - In the goodness-of-fit test of parameters of the multinomial distribution we show that the exact multinomial test is asymptotically equivalent to the likelihood ratio test by using Stirling's formula. In an r×c contingency table, we show that the extended Fisher's exact test conditional on row and column margins for the test of independence is also asymptotically equivalent to the likelihood ratio test. From the Bahadur asymptotic optimality of the likelihood ratio test in both unconditional and conditional cases, we prove that the two exact tests are asymptotically optimal in the sense of Bahadur efficiency.

UR - http://www.scopus.com/inward/record.url?scp=0043098353&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0043098353&partnerID=8YFLogxK

U2 - 10.1016/S0167-7152(99)00010-3

DO - 10.1016/S0167-7152(99)00010-3

M3 - Article

AN - SCOPUS:0043098353

SN - 0167-7152

VL - 44

SP - 201

EP - 207

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 2

ER -

An optimal property of the exact multinomial test and the extended Fisher's exact test

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this