Negative Binomial and Generalized Poisson Regression Model for Death Due to Dengue Hemorrhagic Fever Data

Risnawati Ibnas (Universitas Islam Negeri Alauddin Makassar)
Satriani Satriani (Universitas Islam Negeri Alauddin Makassar)
Khalilah Nurfadilah (Universitas Islam Negeri Alauddin Makassar)

EIGEN MATHEMATICS JOURNAL
Vol. 6 No. 1 Juni 2023

Data on the number of deaths due to Dengue Fever in statistics is count data often approximated by a Poisson distribution. However, if overdispersion occurs, Poisson regression is no longer sufficient, so the Negative Binomial and Generalized Poisson Regression approaches are used. From the two models, the best model was chosen based on the smallest AIC value, 66.50, namely the Negative Binomial Regression model. From this model, factors that have a significant effect are determined based on the p-value, and the factor ratio of health facilities per 100,000 population  is obtained.