Inference on dengue epidemics with Bayesian regime switching models

by Jue Tao Lim, Borame Sue Dickens, Sun Haoyang, Ng Lee Ching, Alex Richard Cook Dengue, a mosquito-borne infectious disease caused by the dengue viruses, is present in many parts of the tropical and subtropical regions of the world. All four serotypes of dengue viruses are endemic in Singapore, an equatorial city-state. Frequent outbreaks occur, sometimes leading to national epidemics. However, few studies have attempted to characterize breakpoints which precede large rises in dengue case counts. In this paper, Bayesian regime switching (BRS) models were employed to infer epidemic and endemic regimes of dengue transmissions, each containing regime specific autoregressiv e processes which drive the growth and decline of dengue cases, estimated using a custom built multi-move Gibbs sampling algorithm. Posterior predictive checks indicate that BRS replicates temporal trends in Dengue transmissions well and nowcast accuracy assessed using a post-hoc classification sche me showed that BRS classification accuracy is robust even under limited data with the AUC-ROC at 0.935. LASSO-based regression and bootstrapping was used to account for plausibly high dimensions of climatic factors affecting Dengue transmissions, which was then estimated using cross-validation to co nduct statistical inference on long-run climatic effects on the estimated regimes. BRS estimates epidemic and endemic regimes of dengue in Singapore which are characterized by persistence across time, lasting an avera...
Source: PLoS Computational Biology - Category: Biology Authors: Source Type: research