Visualization and Negative Binomial Model of Dengue Incidence Rate in the Central Region of Malaysia

Abstract

Dengue fever (DF) is a mosquito-borne viral illness caused by four categories of dengue virus (DENV), DENV-1, DENV-2, DENV-3, and DENV-4, all of which are members of the Flaviviridae family that mostly found in the species of Aedes aegypti and Aedes albopictus species. Humid and warm condition is a suitable condition for the breed of the mosquitoes and their life style. The aim of this study is to visualise the trend of the dengue incidence rate (DIR), to obtain the significant climate variables that related to the DIR and to explore the model that best represents the data. The data of DIR from 2001 to 2009 specifically in the central region of Malaysia; Selangor and Kuala Lumpur were utilised. In achieving the objectives, scatter plot was used to visualise the relationship of the climate variables and DIR while Poisson GLM and Negative Binomial GLM model were used to model the data then statistical test such as AIC, BIC and deviance were utilised for the model selection. The monthly trend of the DIR in Selangor and Kuala Lumpur shows a seasonal trend while Negative Binomial GLM model is the best model for both Selangor and Kuala Lumpur.