Employing ARIMA Forecasting Technique for Dengue Fever Outbreak Prediction in Toba Tek Singh and Jhang District: A Weather-Based Approach

Abstract

In many areas of Pakistan, Dengue is one of the most prevalent and fatal infections. In Pakistan, the frequency and geographic spread of Dengue Fever has grown during the previous ten years. Based on this context, an effort was made to transform the monthly data on Dengue Fever that was available from the two districts of Punjab during the previous four years into a seasonal ARIMA model to predict disease burden. To lower the prevalence of this condition, an accurate forecast-based model is necessary. DHF and other infectious diseases are studied and predicted using time-series methods in epidemiology, such as Autoregressive Integrated Moving Average (ARIMA) models. The current study used a time-series technique to try and predict the monthly verified DHF cases. The information was gathered from the Toba Tek Singh and Jhang District Head Quarter Hospitals and covered the period from March 2019 to December 2022. Several ARIMA models were implemented using Python and a sample data set, and the best seasonal ARIMA model was found. The chosen model was then applied to predict the monthly incidence of Dengue Fever starting in the following years, 2023 and 2024. The results showed that in comparison to any other model, the ARIMA (2,1,2) and ARIMA (2,0,2) models were able to predict the DHF epidemics in both Districts respectively. Using the ARIMA models, we further predicted DHF cases over 19-month intervals beginning in March 2023 and ending in December 2024. The findings revealed large seasonal DHF epidemics, especially from March to September. A substantial seasonal association between DHF and the ambient temperature was rev ealed by the highest instances, which were seen in August and September. In Toba Tek Singh and Jhang City, this study is the first attempt to analyze the time series model for DHF patients and forecast upcoming outbreaks. The research could aid in the development of effective public health policies for disea se detection and management, particularly in the early stages of outbreaks.