Forecasting Typhoid Fever Trend Using Arima Models

Abstract

Typhoid fever remains a significant public health concern globally, with approximately 21 million cases and 200,000 deaths annually. Despite efforts to control the disease, its incidence exhibits complex patterns, influenced by climate variables, sanitation practices, vaccination coverage, population density, and economic factors. The impact of this disease is particularly severe in rural areas, where there’s lack of sanitation hygiene. Preventing this disease in rural areas is significant challenges due to the lack of hospitals and inadequate disease prevention. Time series analysis assist health authorities in anticipating and preparing for potential outbreaks, leading to more effective and timely interventions. This study examined the trend of the reported cases of Typhoid fever. According to data on the reported cases of typhoid fever retrieved from dataset expository called Kaggle used for the analysis. The data space for the period of Nine (9) years, started from January, 2015 to December, 2023. Seasonal ARIMA models were developed for the typhoid fever trend, but ARIMA (2, 0, 0) with non-zero mean was selected as the best performed model based on the AIC value. Furthermore, the results of the ARIMA (2, 0, 0) forecast shows an increased in the cases of the typhoid fever due to the series trend. And this study Develop national guidelines for typhoid fever prevention and control.