Authors :
Herawati, N.; Fitriyani, A.; Saidi, S.; Setiawan, E.; Meliantari, D.
Volume/Issue :
Volume 8 - 2023, Issue 6 - June
Google Scholar :
https://bit.ly/3TmGbDi
Scribd :
https://tinyurl.com/5ca2m93b
DOI :
https://doi.org/10.5281/zenodo.8031706
Abstract :
The Poisson regression model is a statistical
model that can be utilized to analyze the relationship
between independent variable X in the structure of
continuous, discrete or mixed data and dependent
variable Y in the form of discrete data. This model has
several assumptions that must be met, one of which is
that the variance value of the dependent variable must
be equal to the average (equidispersion). If the variance
value is larger than the mean, data overdispersion will
occur. This study will compare the performance of
Quasi-Poisson regression, Zero Inflated Poisson (ZIP)
and Zero Inflated Negative Binomial (ZINB) regression
models on overdispersion data, namely filariasis cases
data in east Lampung province, Indonesia. The results
showed that the ZIP method was the best for modeling
data overdispersion compared to Quasi-Poisson
regression and ZINB method. This is indicated by the
smallest QAIC, QAICc and RMSE values of the ZIP
model when compared to the Quasi-Poisson and ZINB
regression models.
Keywords :
Overdispertion; Quasi-Poisson; zero inflated Poisson; zero inflated negative binomial.
The Poisson regression model is a statistical
model that can be utilized to analyze the relationship
between independent variable X in the structure of
continuous, discrete or mixed data and dependent
variable Y in the form of discrete data. This model has
several assumptions that must be met, one of which is
that the variance value of the dependent variable must
be equal to the average (equidispersion). If the variance
value is larger than the mean, data overdispersion will
occur. This study will compare the performance of
Quasi-Poisson regression, Zero Inflated Poisson (ZIP)
and Zero Inflated Negative Binomial (ZINB) regression
models on overdispersion data, namely filariasis cases
data in east Lampung province, Indonesia. The results
showed that the ZIP method was the best for modeling
data overdispersion compared to Quasi-Poisson
regression and ZINB method. This is indicated by the
smallest QAIC, QAICc and RMSE values of the ZIP
model when compared to the Quasi-Poisson and ZINB
regression models.
Keywords :
Overdispertion; Quasi-Poisson; zero inflated Poisson; zero inflated negative binomial.
