Potential roles of social distancing in mitigating the spread of coronavirus disease 2019 (COVID-19) in South Korea

Sang Woo Park; Kaiyuan Sun; Cécile Viboud; Bryan T Grenfell; Jonathan Dushoff

where f is the negative binomial distribution with time-dependent mean µ(t i ) and dispersion parameter θ. This likelihood accounts for the fact that the delay between symptom onset and confirmation cannot be longer than t max − t i (otherwise, the case will be reported after t max ). Convergence is assessed by the lack of warning messages from the rstan package [23]. For each combination of date of infection and a posterior sample of the forward delay distribution, we drew 1000 samples of incubation periods and onset-to-confirmation delays and calculated the median probability that an individual infected on a given day will be confirmed before March 16, 2020. Finally, we divided the daily number of infected cases by the median probability this probability. We used the reconstructed incidence time series to estimate R t .