Predicting Dengue Outbreaks in Cagayan de Oro, Philippines Using Facebook Prophet and the ARIMA Model for Time Series Forecasting

Abstract

In terms of public health, dengue fever continues to be an important issue. In recent years, there has been a sharp rise in dengue-related cases all throughout the world. The number of reported cases was consistently rising, especially in the cities of the Philippines. Time series analysis is used in this study to forecast the frequency of dengue by modeling the weekly incidence of cases of dengue hemorrhagic fever (DHF) in Cagayan de Oro City, Philippines. This study uses a dataset of confirmed dengue cases that were downloaded from the Department of Health's (DOH) official website to compare the effectiveness and precision of Facebook Prophet's and ARIMA's forecasting models. The performance indicators of the Facebook Prophet and ARIMA approaches are contrasted on the same dataset to choose the best accurate forecast model. The period covered by the dataset chosen for this study runs from the first week of 2016 to the sixteenth week of 2022. The performance of the forecast models is then evaluated by comparing them to the last 66 weeks of actual data. The result of this study shows that Facebook Prophet model outperforms ARIMA model.