Bivariate zero-inflated negative binomial regression model with applications

Pouya Faroughi, Noriszura Ismail

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Count data often display excessive number of zero outcomes than are expected in the Poisson regression model. The zero-inflated Poisson regression model has been suggested to handle zero-inflated data, whereas the zero-inflated negative binomial (ZINB) regression model has been fitted for zero-inflated data with additional overdispersion. For bivariate and zero-inflated cases, several regression models such as the bivariate zero-inflated Poisson (BZIP) and bivariate zero-inflated negative binomial (BZINB) have been considered. This paper introduces several forms of nested BZINB regression model which can be fitted to bivariate and zero-inflated count data. The mean–variance approach is used for comparing the BZIP and our forms of BZINB regression model in this study. A similar approach was also used by past researchers for defining several negative binomial and zero-inflated negative binomial regression models based on the appearance of linear and quadratic terms of the variance function. The nested BZINB regression models proposed in this study have several advantages; the likelihood ratio tests can be performed for choosing the best model, the models have flexible forms of marginal mean–variance relationship, the models can be fitted to bivariate zero-inflated count data with positive or negative correlations, and the models allow additional overdispersion of the two dependent variables.

Original languageEnglish
Pages (from-to)1-21
Number of pages21
JournalJournal of Statistical Computation and Simulation
DOIs
Publication statusAccepted/In press - 29 Jul 2016

Fingerprint

Binomial Model
Negative Binomial
Regression Model
Zero
Count Data
Poisson Regression
Overdispersion
Poisson Model
Negative binomial regression
Regression model
Siméon Denis Poisson
Variance Function
Likelihood Ratio Test

Keywords

  • Bivariate counts
  • healthcare
  • negative binomial
  • zero-inflation

ASJC Scopus subject areas

  • Statistics and Probability
  • Modelling and Simulation
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

Bivariate zero-inflated negative binomial regression model with applications. / Faroughi, Pouya; Ismail, Noriszura.

In: Journal of Statistical Computation and Simulation, 29.07.2016, p. 1-21.

Research output: Contribution to journalArticle

@article{88e7e8d72aa04fd1b413e15c56a852cc,
title = "Bivariate zero-inflated negative binomial regression model with applications",
abstract = "Count data often display excessive number of zero outcomes than are expected in the Poisson regression model. The zero-inflated Poisson regression model has been suggested to handle zero-inflated data, whereas the zero-inflated negative binomial (ZINB) regression model has been fitted for zero-inflated data with additional overdispersion. For bivariate and zero-inflated cases, several regression models such as the bivariate zero-inflated Poisson (BZIP) and bivariate zero-inflated negative binomial (BZINB) have been considered. This paper introduces several forms of nested BZINB regression model which can be fitted to bivariate and zero-inflated count data. The mean–variance approach is used for comparing the BZIP and our forms of BZINB regression model in this study. A similar approach was also used by past researchers for defining several negative binomial and zero-inflated negative binomial regression models based on the appearance of linear and quadratic terms of the variance function. The nested BZINB regression models proposed in this study have several advantages; the likelihood ratio tests can be performed for choosing the best model, the models have flexible forms of marginal mean–variance relationship, the models can be fitted to bivariate zero-inflated count data with positive or negative correlations, and the models allow additional overdispersion of the two dependent variables.",
keywords = "Bivariate counts, healthcare, negative binomial, zero-inflation",
author = "Pouya Faroughi and Noriszura Ismail",
year = "2016",
month = "7",
day = "29",
doi = "10.1080/00949655.2016.1213843",
language = "English",
pages = "1--21",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - Bivariate zero-inflated negative binomial regression model with applications

AU - Faroughi, Pouya

AU - Ismail, Noriszura

PY - 2016/7/29

Y1 - 2016/7/29

N2 - Count data often display excessive number of zero outcomes than are expected in the Poisson regression model. The zero-inflated Poisson regression model has been suggested to handle zero-inflated data, whereas the zero-inflated negative binomial (ZINB) regression model has been fitted for zero-inflated data with additional overdispersion. For bivariate and zero-inflated cases, several regression models such as the bivariate zero-inflated Poisson (BZIP) and bivariate zero-inflated negative binomial (BZINB) have been considered. This paper introduces several forms of nested BZINB regression model which can be fitted to bivariate and zero-inflated count data. The mean–variance approach is used for comparing the BZIP and our forms of BZINB regression model in this study. A similar approach was also used by past researchers for defining several negative binomial and zero-inflated negative binomial regression models based on the appearance of linear and quadratic terms of the variance function. The nested BZINB regression models proposed in this study have several advantages; the likelihood ratio tests can be performed for choosing the best model, the models have flexible forms of marginal mean–variance relationship, the models can be fitted to bivariate zero-inflated count data with positive or negative correlations, and the models allow additional overdispersion of the two dependent variables.

AB - Count data often display excessive number of zero outcomes than are expected in the Poisson regression model. The zero-inflated Poisson regression model has been suggested to handle zero-inflated data, whereas the zero-inflated negative binomial (ZINB) regression model has been fitted for zero-inflated data with additional overdispersion. For bivariate and zero-inflated cases, several regression models such as the bivariate zero-inflated Poisson (BZIP) and bivariate zero-inflated negative binomial (BZINB) have been considered. This paper introduces several forms of nested BZINB regression model which can be fitted to bivariate and zero-inflated count data. The mean–variance approach is used for comparing the BZIP and our forms of BZINB regression model in this study. A similar approach was also used by past researchers for defining several negative binomial and zero-inflated negative binomial regression models based on the appearance of linear and quadratic terms of the variance function. The nested BZINB regression models proposed in this study have several advantages; the likelihood ratio tests can be performed for choosing the best model, the models have flexible forms of marginal mean–variance relationship, the models can be fitted to bivariate zero-inflated count data with positive or negative correlations, and the models allow additional overdispersion of the two dependent variables.

KW - Bivariate counts

KW - healthcare

KW - negative binomial

KW - zero-inflation

UR - http://www.scopus.com/inward/record.url?scp=84980049318&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84980049318&partnerID=8YFLogxK

U2 - 10.1080/00949655.2016.1213843

DO - 10.1080/00949655.2016.1213843

M3 - Article

SP - 1

EP - 21

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

ER -