State Estimation of the Spread of COVID-19 in Saudi Arabia using Extended Kalman Filter

Abstract

COVID-19 has caused global concern as the World Health Organization (WHO) considered it a global pandemic that has affected all countries to different extent. Numerous studies have examined the behaviour of the pandemic using a wide variety of mathematical models. In this paper, we consider the nonlinear compartmental epidemiological dynamical system model in the Susceptible-Exposed-Infected-Quarantined-Recovered-Deceased (SEIQRD) form based on the recursive estimator known as the extended Kalman filter (EKF) to predict the evolution of the COVID-19 pandemic in Saudi Arabia. We adopt the nested sampling algorithm for parameter estimation and uncertainty quantification of the SEIQRD model parameters using real data. Our simulation results show that the EKF can not only predict the evolution of the directly measured variables i.e. the total death (D) and active case (I) but can also be useful in the estimation of the unmeasurable state variables and help predicting their future trends.