Elementary time-delay dynamics of COVID-19 disease

Jose Menendez

maximum. An important characteristic of the data is that the ratio is very small for both countries (20% in the case of China and 5% in the case of South Korea). To reproduce these observed ratios within SEIR models, very long mean infection periods are required, exceeding 100 days. Such long infection periods are unrealistic and lead to a t > tpeak decrease in the number of infected people that is much slower than observed. The discrepancy between SEIR models and the available data becomes even worse if one considers the fact that the time lag for reporting a recovered person is likely to be much shorter than the incubation time for the disease. Therefore, the "instantaneous" predicted quantities and from the models do not correspond to the observed and but to "corrected" quantities that can approximated as , (1) where T is the mean incubation time. Using T = 7 days, in line with a reported incubation time between 5 and 11 days [6] , one obtains the data in Fig. 1 and Fig. 2 . In the case of China, for example, the ratio has now become as small as 8%.