Dengue Spread Model using Climate Variables

Many cases related to dengue reported each year, threatening the life of certain victims. The number of victims increase annually and become biggest issue in early 21st century until today. Therefore, the project takes initiative to develop model that can predicts dengue outbreak. This project aims to deliver the early warning system for possible dengue outbreak, thus enhance the efficiency of primary dengue surveillance system. The forecasting model using Autoregressive Integrated Moving Average (ARIMA) is generated using the data acquired from Ministry of Health Malaysia which consisted of data from January 2010 to December 2011 in Selangor. Selangor is chosen because of the rising in case of dengue and has the most population compared to other state. The model is then validated using the data from January 2012 to June 2012. The relationship between dengue incidence with climate variable is examined using cross correlations to reducing the error forecasted. The result of this study revealed that ARIMA(1,1,2)(1,1,0)20, ARIMA(1,0,1)(1,1,0)20, ARIMA(0,1,1)(1,1,1)20, ARIMA(1,0,1)(1,1,1)20, ARIMA(1,0,2)(0,1,1)20, ARIMA(1,0,2)(0,1,1)20 and ARIMA(1,0,1)(0,1,1)20 is respectively the result for Petaling, Gombak, Klang, Kuala Selangor, Hulu Langat, Hulu Selangor and Sepang.