Score test for testing zero-inflated Poisson regression against zero-inflated generalized Poisson alternatives

Hossein Zamani, Noriszura Ismail

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

In several cases, count data often have excessive number of zero outcomes. This zero-inflated phenomenon is a specific cause of overdispersion, and zero-inflated Poisson regression model (ZIP) has been proposed for accommodating zero-inflated data. However, if the data continue to suggest additional overdispersion, zero-inflated negative binomial (ZINB) and zero-inflated generalized Poisson (ZIGP) regression models have been considered as alternatives. This study proposes the score test for testing ZIP regression model against ZIGP alternatives and proves that it is equal to the score test for testing ZIP regression model against ZINB alternatives. The advantage of using the score test over other alternative tests such as likelihood ratio and Wald is that the score test can be used to determine whether a more complex model is appropriate without fitting the more complex model. Applications of the proposed score test on several datasets are also illustrated.

Original languageEnglish
Pages (from-to)2056-2068
Number of pages13
JournalJournal of Applied Statistics
Volume40
Issue number9
DOIs
Publication statusPublished - Sep 2013

Fingerprint

Poisson Regression
Score Test
Siméon Denis Poisson
Testing
Alternatives
Zero
Regression Model
Overdispersion
Negative Binomial
Poisson Model
Score test
Poisson regression
Count Data
Likelihood Ratio
Regression model
Continue

Keywords

  • overdispersion
  • score test
  • zero-inflated generalized Poisson regression
  • zero-inflated Poisson regression
  • zero-inflation

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Score test for testing zero-inflated Poisson regression against zero-inflated generalized Poisson alternatives. / Zamani, Hossein; Ismail, Noriszura.

In: Journal of Applied Statistics, Vol. 40, No. 9, 09.2013, p. 2056-2068.

Research output: Contribution to journalArticle

@article{db210180faff4a098f894746f1347536,
title = "Score test for testing zero-inflated Poisson regression against zero-inflated generalized Poisson alternatives",
abstract = "In several cases, count data often have excessive number of zero outcomes. This zero-inflated phenomenon is a specific cause of overdispersion, and zero-inflated Poisson regression model (ZIP) has been proposed for accommodating zero-inflated data. However, if the data continue to suggest additional overdispersion, zero-inflated negative binomial (ZINB) and zero-inflated generalized Poisson (ZIGP) regression models have been considered as alternatives. This study proposes the score test for testing ZIP regression model against ZIGP alternatives and proves that it is equal to the score test for testing ZIP regression model against ZINB alternatives. The advantage of using the score test over other alternative tests such as likelihood ratio and Wald is that the score test can be used to determine whether a more complex model is appropriate without fitting the more complex model. Applications of the proposed score test on several datasets are also illustrated.",
keywords = "overdispersion, score test, zero-inflated generalized Poisson regression, zero-inflated Poisson regression, zero-inflation",
author = "Hossein Zamani and Noriszura Ismail",
year = "2013",
month = "9",
doi = "10.1080/02664763.2013.804904",
language = "English",
volume = "40",
pages = "2056--2068",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",
number = "9",

}

TY - JOUR

T1 - Score test for testing zero-inflated Poisson regression against zero-inflated generalized Poisson alternatives

AU - Zamani, Hossein

AU - Ismail, Noriszura

PY - 2013/9

Y1 - 2013/9

N2 - In several cases, count data often have excessive number of zero outcomes. This zero-inflated phenomenon is a specific cause of overdispersion, and zero-inflated Poisson regression model (ZIP) has been proposed for accommodating zero-inflated data. However, if the data continue to suggest additional overdispersion, zero-inflated negative binomial (ZINB) and zero-inflated generalized Poisson (ZIGP) regression models have been considered as alternatives. This study proposes the score test for testing ZIP regression model against ZIGP alternatives and proves that it is equal to the score test for testing ZIP regression model against ZINB alternatives. The advantage of using the score test over other alternative tests such as likelihood ratio and Wald is that the score test can be used to determine whether a more complex model is appropriate without fitting the more complex model. Applications of the proposed score test on several datasets are also illustrated.

AB - In several cases, count data often have excessive number of zero outcomes. This zero-inflated phenomenon is a specific cause of overdispersion, and zero-inflated Poisson regression model (ZIP) has been proposed for accommodating zero-inflated data. However, if the data continue to suggest additional overdispersion, zero-inflated negative binomial (ZINB) and zero-inflated generalized Poisson (ZIGP) regression models have been considered as alternatives. This study proposes the score test for testing ZIP regression model against ZIGP alternatives and proves that it is equal to the score test for testing ZIP regression model against ZINB alternatives. The advantage of using the score test over other alternative tests such as likelihood ratio and Wald is that the score test can be used to determine whether a more complex model is appropriate without fitting the more complex model. Applications of the proposed score test on several datasets are also illustrated.

KW - overdispersion

KW - score test

KW - zero-inflated generalized Poisson regression

KW - zero-inflated Poisson regression

KW - zero-inflation

UR - http://www.scopus.com/inward/record.url?scp=84883622602&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883622602&partnerID=8YFLogxK

U2 - 10.1080/02664763.2013.804904

DO - 10.1080/02664763.2013.804904

M3 - Article

VL - 40

SP - 2056

EP - 2068

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

IS - 9

ER -