A discrete stochastic model of the COVID-19 outbreak: Forecast and control

Math Biosci Eng. 2020 Mar 16;17(4):2792-2804. doi: 10.3934/mbe.2020153.

Abstract

The novel Coronavirus (COVID-19) is spreading and has caused a large-scale infection in China since December 2019. This has led to a significant impact on the lives and economy in China and other countries. Here we develop a discrete-time stochastic epidemic model with binomial distributions to study the transmission of the disease. Model parameters are estimated on the basis of fitting to newly reported data from January 11 to February 13, 2020 in China. The estimates of the contact rate and the effective reproductive number support the efficiency of the control measures that have been implemented so far. Simulations show the newly confirmed cases will continue to decline and the total confirmed cases will reach the peak around the end of February of 2020 under the current control measures. The impact of the timing of returning to work is also evaluated on the disease transmission given different strength of protection and control measures.

Keywords: COVID-19; control measures; data fitting; parameter estimation; stochastic model.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Basic Reproduction Number / statistics & numerical data
  • Betacoronavirus*
  • COVID-19
  • China / epidemiology
  • Computer Simulation
  • Coronavirus Infections / epidemiology*
  • Coronavirus Infections / prevention & control
  • Coronavirus Infections / transmission
  • Humans
  • Mathematical Concepts
  • Models, Biological*
  • Pandemics* / prevention & control
  • Pandemics* / statistics & numerical data
  • Pneumonia, Viral / epidemiology*
  • Pneumonia, Viral / prevention & control
  • Pneumonia, Viral / transmission
  • SARS-CoV-2
  • Stochastic Processes