Modeling Dengue Fever Cases in West Java Using Gaussian, Poisson, and Negative Binomial Regression

Dengue Hemorrhagic Fever (DHF) remains a major health problem in West Java Province, which has a tropical climate and dense population. This study aims to model DHF cases using Gaussian, Poisson, and Negative Binomial regression. The data used include the number of DHF cases in 2024 as the response variable and eight predictor variables, namely humidity, amount of rainfall, real expenditure per capita, poverty ratio, population density, percentage of households with clean and healthy lifestyles, ratio of health facilities, and average height of the area above sea level (ASL) according to rainfall stations. Data were obtained from the Central Statistics Agency (BPS) and the West Java Provincial Health Office. The analysis results showed that Poisson and Gaussian regression were less suitable because they were unable to handle overdispersion in the data. The Negative Binomial model provided the best results with the smallest AIC value. Two variables that significantly influenced the number of dengue cases were population density (variable 5) and the ratio of health facilities (variable 7)