论文标题
大都会级Covid-19爆发：它们有多相似？
Metropolitan-scale COVID-19 outbreaks: how similar are they?
论文作者
论文摘要
在这项研究中，我们从2020年1月21日至3月25日开始使用美国县级COVID-19案例数据来研究大都市量表的案例增长的指数行为。特别是，我们假设所有局部暴发都处于早期阶段（在病例数中经历指数增长，或有效地包含），并比较了不同大都市地区的不同简单指数和线性增长模型的解释性性能。尽管我们发现城市规模和指数增长率之间没有关系（与$ r0 $直接相关，这表示感染个体感染的平均病例数量），但我们确实发现，较大的城市似乎开始提早扩散，因此在提交时大流行的阶段更高级阶段。我们还使用更多的数据来计算鉴于模型的预测错误，并发现在许多城市中，在3月26日之前对数据进行训练的指数增长模型在这一较新的时期（3月26日至30日）的案件数量很差，这可能表明减少了通过社会疏远来促进的新案例数量。
In this study, we use US county-level COVID-19 case data from January 21-March 25, 2020 to study the exponential behavior of case growth at the metropolitan scale. In particular, we assume that all localized outbreaks are in an early stage (either undergoing exponential growth in the number of cases, or are effectively contained) and compare the explanatory performance of different simple exponential and linear growth models for different metropolitan areas. While we find no relationship between city size and exponential growth rate (directly related to $R0$, which denotes average the number of cases an infected individual infects), we do find that larger cities seem to begin exponential spreading earlier and are thus in a more advanced stage of the pandemic at the time of submission. We also use more recent data to compute prediction errors given our models, and find that in many cities, exponential growth models trained on data before March 26 are poor predictors for case numbers in this more recent period (March 26-30), likely indicating a reduction in the number of new cases facilitated through social distancing.
