UCLA

This paper investigates whether the distribution of SARS-CoV-2 (COVID-19) transmission times can be reliably estimated using only case count data, employing the Hawkes model as the analytical framework. Hawkes point processes, widely recognized for modeling and analyzing time-to-event data, offer a robust approach to understanding transmission dynamics. This study fits the Hawkes model with varying productivity levels to case count data from all 50 U.S. states. Transmission time density is estimated using nonparametric methods and normal approximations. The findings indicate that, for most states, the mean transmission time is approximately 7 days, with a standard deviation of about 1 day (Science, 2020). These estimates are compared across states and with prior reports, revealing slightly shorter average transmission durations and reduced variability in this study. Furthermore, the results highlight that the virus can be transmitted as early as the first day of contact, emphasizing its potential for rapid spread (World Health Organization, 2020). As derived from this analysis, a deeper understanding of SARS-CoV-2 transmission dynamics carries significant implications for public health modeling and policy-making (Pan et al., 2020).