Time series patterns of dengue and associated climate variables in Bangladesh and Singapore (2000-2020): a comparative study of statistical models to forecast dengue cases.

Abstract

This study explores the effect of climate factors on dengue incidence in Bangladesh and Singapore from 2000 to 2020. Various forecasting models, including Seasonal ARIMA, Poisson regression, artificial neural networks (ANN), support vector machines (SVM), and other statistical techniques, were applied to forecast dengue trends and generate future datasets. The results showed that Bangladesh has higher mean temperatures. The Poisson regression indicated that rainfall positively correlated with dengue cases in Bangladesh, while humidity and sunshine showed negative associations. In Singapore, temperature was positively associated with dengue cases, while rainfall and humidity had inverse relationships. ARIMA models predicted Singapore would experience the highest dengue cases by 2023. Based on the Root Mean Square Error (RMSE), the TBATS model was most accurate for predicting dengue cases in Bangladesh, while both ARIMA and TBATS models performed well in Singapore. This study provides valuable insights for policymakers in Singapore and Bangladesh to develop climate-based warning systems.