中文

Operations Research Transactions ›› 2020, Vol. 24 ›› Issue (1): 1-12.doi: 10.15960/j.cnki.issn.1007-6093.2020.01.001

    Next Articles

A probability model for estimating the expected number of the newly infected and predicting the trend of the diagnosed

DING Zhiwei1, LIU Yanyun2, KONG Jing1, ZHANG Hong1, ZHANG Yi3, DAI Yuhong4, YANG Zhouwang1,*   

  1. 1. School of Mathematical Sciences, University of Science and Technology of China, Hefei 230026, China;
    2. School of Mathematical Sciences, Peking University, Beijing 100871, China;
    3. Beijing Institute of Big Data Research, Beijing 100871, China;
    4. Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China
  • Received:2020-03-02 Published:2020-03-09

PDF

954

Abstract: After 2019 novel cornavirus disease (COVID-19) appeared in Wuhan in early December 2019, it broke out in mid-to-late January 2020 and quickly spread throughout the country. So far, it has spread in dozens of countries and regions, the scientific and efficient understanding of epidemic development is essential for prevention and control. The number of infected people is a key indicator for assessing the situation of the epidemic, helping decision-makers formulate policies in time. This paper uses the maximum likelihood estimation method to obtain estimators of the number of newly infected people across the country except Hubei province. Moreover, Bootstrap simulation enables us to obtain confidence intervals for the estimators. Based on these solutions of the model, we further calculate the number of existing infected but undiagnosed people and predict the trend of the newly diagnosed for the next few days, providing suggestions on returning to work.

Key words: 2019 novel coronavirus, number of the infected, maximum likelihood estimation, EM algorithm, Bootstrap

CLC Number: