Climate Factors Improve Accuracy of Time Series Model for Dengue Hemorrhagic Fever Forecasting in Pasuruan Regency, Indonesia Sari Qurrotul Aini Meta Puspita1,2, Notobroto Hari Basuki3, Mahmudah3 1Student Master of Public Health at Airlangga University, Surabaya, Indonesia 2Research Center for Humanities and Health Management, Ministry of Health, Indonesia 3Departments of Biostatistics, Faculty of Public Health, Airlangga University, Surabaya, Indonesia *Corresponding Author: Qurrotul Aini Meta Puspita Sari Student Master of Public Health, Airlangga University, Surabaya, Indonesia, 60115, Email: qurrotul.aini.meta-2017@fkm.unair.ac.id
Online published on 23 December, 2019. Abstract A large number of dengue cases in Indonesia must be followed by vector control, and efforts to terminate the chain of disease transmission with promotive and preventive activity. Pasuruan Regency is one of the regions in East Java, Indonesia that has a large number of dengue cases and exists monthly throughout the year. Accurate forecasting of Dengue Hemorrhagic Fever (DHF) cases before the outbreak in Pasuruan, can help public health practitioners to prioritize public health activities. This study aims to obtain a time series forecasting model for the number of dengue cases with the addition of climate factor predictors as rainfall, air humidity, air temperature and duration of irradiation. Secondary data of DHF and climate factor in 2015–2017 were analyzed by ARIMA and Multi Input Transfer Function. Forecast model was evaluate by actual data in 2018. The Result showed that he addition of the climate predictor factor as rainfall, air humidity and air temperature in Multi Input Transfer Function model succeeded in increasing forecasting accuracy compared to the ARIMA (1, 1, 1). RMSE Transfer Function model decrease by 9.19%. DHF was significantly affected by DHF in the previous month, rainfall in the previous 1 and 2 months, air humidity 3 and 4 months before and the air temperature 1 and 2 months before and interaction of random effects of rainfall, humidity and air temperature with DHF one month earlier. Top Keywords DHF, Forecasting, ARIMA, Transfer Function, Climate. Top |