Approximate distribution and test of fit for the clustering effect in the dirichlet multinomial model

Jeffrey Wilson, Jeffrey R. Wilson

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The Dirichlet-multinomial model is considered as a model for cluster sampling. The model assumes that the design's covariance matrix is a constant times the covariance under multinomial sampling. The use of this model requires estimating a parameter C, that measures the clustering effect. In this paper, a regression estimate for C is obtained. An approximate distribution of this estimator is obtained through the use of asymptotic techniques. A goodness of fit statistic for testing the fit of the Dirichlet Multinomial model is also obtained, based on those asymptotic techniques. These statistics provide a means of knowing when the data satisfy the model assumption. These results are used to analyze data concerning the authorship of Greek prose.

Original languageEnglish (US)
Pages (from-to)1235-1249
Number of pages15
JournalCommunications in Statistics - Theory and Methods
Volume15
Issue number4
DOIs
StatePublished - Jan 1 1986

Fingerprint

Multinomial Model
Dirichlet
Clustering
Cluster Sampling
Regression Estimate
Goodness of fit
Time Constant
Model
Covariance matrix
Statistic
Statistics
Estimator
Testing

Keywords

  • Asymptotic
  • cluster sampling
  • goodness-of-fit
  • regression

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Approximate distribution and test of fit for the clustering effect in the dirichlet multinomial model. / Wilson, Jeffrey; Wilson, Jeffrey R.

In: Communications in Statistics - Theory and Methods, Vol. 15, No. 4, 01.01.1986, p. 1235-1249.

Research output: Contribution to journalArticle

@article{a4e1293dfa1b4cda9f8864aefc25fd1b,
title = "Approximate distribution and test of fit for the clustering effect in the dirichlet multinomial model",
abstract = "The Dirichlet-multinomial model is considered as a model for cluster sampling. The model assumes that the design's covariance matrix is a constant times the covariance under multinomial sampling. The use of this model requires estimating a parameter C, that measures the clustering effect. In this paper, a regression estimate for C is obtained. An approximate distribution of this estimator is obtained through the use of asymptotic techniques. A goodness of fit statistic for testing the fit of the Dirichlet Multinomial model is also obtained, based on those asymptotic techniques. These statistics provide a means of knowing when the data satisfy the model assumption. These results are used to analyze data concerning the authorship of Greek prose.",
keywords = "Asymptotic, cluster sampling, goodness-of-fit, regression",
author = "Jeffrey Wilson and Wilson, {Jeffrey R.}",
year = "1986",
month = "1",
day = "1",
doi = "10.1080/03610928608829180",
language = "English (US)",
volume = "15",
pages = "1235--1249",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - Approximate distribution and test of fit for the clustering effect in the dirichlet multinomial model

AU - Wilson, Jeffrey

AU - Wilson, Jeffrey R.

PY - 1986/1/1

Y1 - 1986/1/1

N2 - The Dirichlet-multinomial model is considered as a model for cluster sampling. The model assumes that the design's covariance matrix is a constant times the covariance under multinomial sampling. The use of this model requires estimating a parameter C, that measures the clustering effect. In this paper, a regression estimate for C is obtained. An approximate distribution of this estimator is obtained through the use of asymptotic techniques. A goodness of fit statistic for testing the fit of the Dirichlet Multinomial model is also obtained, based on those asymptotic techniques. These statistics provide a means of knowing when the data satisfy the model assumption. These results are used to analyze data concerning the authorship of Greek prose.

AB - The Dirichlet-multinomial model is considered as a model for cluster sampling. The model assumes that the design's covariance matrix is a constant times the covariance under multinomial sampling. The use of this model requires estimating a parameter C, that measures the clustering effect. In this paper, a regression estimate for C is obtained. An approximate distribution of this estimator is obtained through the use of asymptotic techniques. A goodness of fit statistic for testing the fit of the Dirichlet Multinomial model is also obtained, based on those asymptotic techniques. These statistics provide a means of knowing when the data satisfy the model assumption. These results are used to analyze data concerning the authorship of Greek prose.

KW - Asymptotic

KW - cluster sampling

KW - goodness-of-fit

KW - regression

UR - http://www.scopus.com/inward/record.url?scp=0039714348&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0039714348&partnerID=8YFLogxK

U2 - 10.1080/03610928608829180

DO - 10.1080/03610928608829180

M3 - Article

VL - 15

SP - 1235

EP - 1249

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 4

ER -