Aswi, Cramb Susanna, White Gentry, Hu Wenbiao, Mengersen Kerrie
Dengue fever has become a major public health problem in several countries. This paper aims to review and compare a number of temporal modeling approaches that have been proposed for predicting or forecasting the occurrence of dengue fever. This review also examines influential covariates considered in these studies. A comprehensive literature search was undertaken in September 2018, using Medline (via Ebscohost), ProQuest, Scopus, and Web of Science electronic databases. The search was confined to articles in English, published in refereed journals between January 2000 and September 2018. The most popular approach to temporal modeling of dengue was found to be an autoregressive integrated moving average (ARIMA) model. A limited number of studies applied Bayesian hierarchical dynamic generalized linear models. Climatic variables were most commonly associated with dengue incidence for temporal modeling. © 2019 Trans Tech Publications Ltd, Switzerland. All rights reserved.
School of Mathematical Sciences, Queensland University of Technology, Brisbane, Australia; Universitas Negeri Makassar, Indonesia; Cancer Council Queensland, Brisbane, Australia; School of Public Health and Social Work, Queensland University of Technology, Australia
