Acta Scientifica Malaysia (ASM)

Traditional infectious disease models often emphasize central tendencies, such as average case counts, while overlooking the importance of time-varying volatility in incidence patterns. This study addresses that gap by investigating the heteroskedastic nature of epidemic data using GARCH-family models. The objective is to evaluate the suitability of GARCH (1,1), EGARCH(1,1), and TGARCH(1,1) models in capturing the dynamic and clustered volatility observed in monthly incidence rates of influenza (2010–2022) and COVID-19 (2020–2022). Disease data were sourced from the World Health Organization (WHO) and the Centers for Disease Control and Prevention (CDC), and were analyzed using descriptive statistics, time series visualization, and maximum likelihood estimation. The findings reveal significant volatility clustering in both diseases, with COVID-19 exhibiting greater asymmetry and sharper spikes. The EGARCH model best captured COVID-19’s asymmetric volatility, while TGARCH was better suited to modeling extreme seasonal peaks in influenza. This study fills a critical gap in the literature by extending volatility modeling—traditionally confined to finance—into epidemiology, where it remains underutilized. The research contributes a novel methodological framework for integrating conditional variance analysis into public health surveillance, thereby enhancing early warning systems, epidemic forecasting, and strategic resource allocation during volatile outbreak periods.