Poisson Regression Modeling Case Study Dengue Fever in Medan City in 2019

Abstract

Dengue Hemorrhagic Fever (DHF) is an infectious disease caused by the dengue virus carried by the Aedes aegypti or Aedes albopictus mosquito which is spread in Southeast Asia. Medan City is one of the endemic areas for dengue fever in North Sumatra Province. This study aims to model the variable cases of dengue fever and determine the factors that have a significant effect on cases of dengue fever in the city of Medan. The method used in modeling the DHF case variable is the Poisson regression method with the response variable (Y) namely the number of DHF cases in Medan City, while the predictor variables are population density, number of health workers, number of health facilities, area height, and average waste production. In Poisson regression analysis, the response variable (Y) must meet the assumption of equidispersion. However, the assumption is often violated, namely overdispersion. Then Negative Binomial Regression was chosen as a non-linear model derived from the Poisson-gamma mixture distribution which is the application of the Generalized Linear Model (GLM) which describes the relationship between the response variable (Y) and the predictor variable (X).