ABSTRACT
The emerging virus, COVID-19, has caused a massive outbreak worldwide. Based on the publicly available contact-tracing data, we identified 509 transmission chains from 20 provinces in China and estimated the serial interval (SI) and generation interval (GI) of COVID-19 in China. Inspired by different possible values of the time-varying reproduction number for the imported cases and the local cases in China, we divided all transmission events into three subsets: imported (the zeroth generation) infecting 1st-generation locals, 1st-generation locals infecting 2nd-generation locals, and other transmissions among 2+. The corresponding SI (GI) is respectively denoted as SI10 ( GI10 ), SI21 ( GI21 ), and SI3+2+ ( GI3+2+ ). A Bayesian approach with doubly interval-censored likelihood is employed to fit the distribution function of the SI and GI. It was found that the estimated SI10=6.52 (95% CI:5.96-7.13) , SI21=6.01 (95%CI:5.44-6.64) , SI3+2+=4.39 (95% CI:3.74-5.15) , and GI10=5.47 (95% CI:4.57-6.45) , GI21=5.01 (95% CI:3.58-7.06) , GI3+2+=4.25 (95% CI:2.82-6.23) . Thus, overall both SI and GI decrease when generation increases.