Estimation of the epidemic properties of the 2019 novel coronavirus: A mathematical modeling study

Jinghua Li; Yijing Wang; Stuart Gilmour; Mengying Wang; Daisuke Yoneoka; Ying Wang; Xinyi You; Jing Gu; Chun Hao; Liping Peng; Zhicheng Du; Dong Roman Xu; Yuantao Hao

This study has several limitations. It was based on confirmed cases, and by excluding suspected cases or mild cases may have under-estimated the rate of spread of the disease. We did not estimate the values of the parameters defining the transition rate from exposed to infectious, or infected to recovered, but fixed them at previously published values. This was a necessary decision because the clinical features of the disease are not yet fully understood, and may affect estimates. However, our intuition after fitting these models is that the maximum likelihood estimate of the force of infection naturally adjusts to fit the value of the recovery rate, and produces a broadly similar value of the basic reproduction number as a result. Furthermore, to adjust for the still-arbitrary nature of these estimates of key parameters, we used some methods that do not depend on any assumptions about these aspects of the disease process. Another limitation of this study is the uncertainty introduced by the use of multiple modeling methods, which we combined with a weighted average. However, we believe that by presenting varying methods with different assumptions along with a weighted average, we enable researchers and policy-makers to make judgements about the dangers of the epidemic without relying on any particular set of assumptions about a disease that is not yet well understood.