Faculty of Science, Urmia University of Technology, Urmia, Iran.
10.22034/cmde.2025.67639.3231
Abstract
This manuscript investigates the use of stochastic differential equations alongside Kalman filtering methodologies within the SIRD framework to enhance the predictive accuracy of infectious disease epidemic modeling. Leveraging empirical data from the COVID-19 pandemic, the study incorporates stochasticity into transmission dynamics by introducing perturbations in the infection rate. A comparative analysis of both deterministic and stochastic versions of the model is presented. The extended Kalman filter is employed to infer system states from noisy observations, enabling real-time epidemic monitoring. Simulation results show that applying the Kalman filter significantly improves predictive accuracy and model fidelity, closely aligning with the simulated data. This integrated modeling approach offers a robust framework for public health planning, providing more reliable forecasts and supporting the timely implementation of intervention measures.
Hajizadeh, J. and Nabati, P. (2026). Forecasting Disease Spread Using the Stochastic Model and Kalman Filter. Computational Methods for Differential Equations, (), -. doi: 10.22034/cmde.2025.67639.3231
MLA
Hajizadeh, J. , and Nabati, P. . "Forecasting Disease Spread Using the Stochastic Model and Kalman Filter", Computational Methods for Differential Equations, , , 2026, -. doi: 10.22034/cmde.2025.67639.3231
HARVARD
Hajizadeh, J., Nabati, P. (2026). 'Forecasting Disease Spread Using the Stochastic Model and Kalman Filter', Computational Methods for Differential Equations, (), pp. -. doi: 10.22034/cmde.2025.67639.3231
CHICAGO
J. Hajizadeh and P. Nabati, "Forecasting Disease Spread Using the Stochastic Model and Kalman Filter," Computational Methods for Differential Equations, (2026): -, doi: 10.22034/cmde.2025.67639.3231
VANCOUVER
Hajizadeh, J., Nabati, P. Forecasting Disease Spread Using the Stochastic Model and Kalman Filter. Computational Methods for Differential Equations, 2026; (): -. doi: 10.22034/cmde.2025.67639.3231
)