A New Zero–Inflated Regression Model with Applications to Australian Health Survey and Biochemistry Graduate Students Data

In this study, we propose a new zero‐inflated regression model as an alternative to zero‐inflated regression models, such as the zero‐inflated Poisson, zero‐inflated negative binomial, zero‐inflated hurdle‐Poisson, and zero‐inflated hurdle negative binomial models. In this regard, we take benefit of...

Full description

Saved in:
Bibliographic Details
Published inJournal of mathematics (Hidawi) Vol. 2024; no. 1
Main Authors Tanış, Caner, Mansour, Mahmoud M., Abd Elrazik, Enayat M., Chesneau, Christophe, Al-Mofleh, Hazem, Afify, Ahmed Z.
Format Journal Article
LanguageEnglish
Published Cairo Hindawi Limited 01.01.2024
Subjects
Datasets
Goodness of fit
Health surveys
Maximum likelihood method
Monte Carlo simulation
Normal distribution
Parameter estimation
Random variables
Regression models
Software
Online AccessGet full text

Cover

More Information
Summary:In this study, we propose a new zero‐inflated regression model as an alternative to zero‐inflated regression models, such as the zero‐inflated Poisson, zero‐inflated negative binomial, zero‐inflated hurdle‐Poisson, and zero‐inflated hurdle negative binomial models. In this regard, we take benefit of the flexibility of the Poisson–Bilal distribution and some of its notable properties. More concretely, it is employed as the baseline distribution to generate a new regression model called the zero‐inflated Poisson‐Bilal regression model. It is designed to be a good alternative for modeling overdispersed data quite effectively. This aspect is emphasized using two real‐world data sets from the medicine and education fields. Furthermore, these data sets are analyzed to compare the goodness‐of‐fit of the suggested zero‐inflated regression model with some of its direct competitors.
ISSN:2314-4629
2314-4785
DOI:10.1155/2024/4081833