A Score Test for Testing a Marginalized Zero-Inflated Poisson Regression Model Against a Marginalized Zero-Inflated Negative Binomial Regression Model

İnan, Gül; Preisser, John S.; Das, Kalyan

doi:10.1007/s13253-017-0314-5

Cited by 4 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As in the comparison of standard zero-inflated models, care must be taken in applying likelihood ratio tests as test statistics may not follow conventional chisquare distributions (Böhning, 1998). A score test for comparing MZIP to MZINB models having a common covariate structure has recently been developed (Inan et al , 2017). Statistical tests for zero-inflation (Ugarte et al , 2004) relative to MZIP and MZINB models could also be developed.…”

Section: Discussionmentioning

confidence: 99%

Marginalized zero‐inflated Poisson models with missing covariates

et al. 2018

Self Cite

View full text Add to dashboard Cite

Unlike zero-inflated Poisson regression, marginalized zero-inflated Poisson (MZIP) models for counts with excess zeros provide estimates with direct interpretations for the overall effects of covariates on the marginal mean. In the presence of missing covariates, MZIP and many other count data models are ordinarily fitted using complete case analysis methods due to lack of appropriate statistical methods and software. This article presents an estimation method for MZIP models with missing covariates. The method, which is applicable to other missing data problems, is illustrated and compared with complete case analysis by using simulations and dental data on the caries preventive effects of a school-based fluoride mouthrinse program.

show abstract

Section: Discussionmentioning

confidence: 99%

Marginalized zero‐inflated Poisson models with missing covariates

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…As Godwin () states as

α \to \infty

(or as

ϕ = \frac{1}{α} \to 0

), the OIZTNB distribution converges to the OIPP distribution. Having some expertise in developing a score test for comparing a marginalized Poisson regression model with its negative binomial counterpart (please, see Inan, Preisser, and Das, ), we would suggest to reparametrize the OIZTNB distribution in terms of over‐dispersion parameter ϕ rather than α. Then testing the OIPP regression model against the OIZTNB regression model can be possible via testing

H_{0} : ϕ = 0

against

H_{A} : ϕ > 0

.…”

Section: Specification Testing Of the Oiztnb Regression Modelmentioning

confidence: 99%

“…We kindly invite the readers to have a look at supplementary file of Inan et al. () for further information on developing a score test.…”

Section: Specification Testing Of the Oiztnb Regression Modelmentioning

confidence: 99%

“…For that reason, a regression model based on a NB distribution may not be a good fit for this kind of data. Inan, Preisser, and Das, 2018), we would suggest to reparametrize the OIZTNB distribution in terms of over-dispersion parameter rather than . Then testing the OIPP regression model against the OIZTNB regression model can be possible via testing 0 ∶ = 0 against ∶ > 0.…”

Section: Illustrative Examplesmentioning

confidence: 99%

“…But, it is also worthy to say that since the evaluation will be under the OIPP distribution, one may not need the derivations of the expectation and variance of OIZTNB distribution that recently became a research topic (Shonkwiler, 2016). We kindly invite the readers to have a look at supplementary file of Inan et al (2018) for further information on developing a score test.…”

Section: Illustrative Examplesmentioning

confidence: 99%

See 2 more Smart Citations

One‐inflation and unobserved heterogeneity in population size estimation by Ryan T. Godwin

İnan

2018

Biometrical J

Self Cite

View full text Add to dashboard Cite

In this study, we would like to show that the one-inflated zero-truncated negative binomial (OIZTNB) regression model can be easily implemented in R via built-in functions when we use mean-parameterization feature of negative binomial distribution to build OIZTNB regression model. From the practitioners' point of view, we believe that this approach presents a computationally convenient way for implementation of the OIZTNB regression model.

show abstract