Dayaratna Kevin D, Gonshorowski Drew, Kolesar Mary
Center for Data Analysis, The Heritage Foundation, Washington, DC, USA.
Mathematics Department, Harvard University, Cambridge, MA, USA.
J Appl Stat. 2022 May 16;50(11-12):2663-2680. doi: 10.1080/02664763.2022.2069232. eCollection 2023.
We examine the impact of economic, demographic, and mobility-related factors have had on the transmission of COVID-19 in 2020. While many models in the academic literature employ linear/generalized linear models, few contributions exist that incorporate spatial analysis, which is useful for understanding factors influencing the proliferation of the disease before the introduction of vaccines. We utilize a Poisson generalized linear model coupled with a spatial autoregressive structure to do so. Our analysis yields a number of insights including that, in some areas of the country, the counterintuitive result that staying at home can lead to increased disease proliferation. Additionally, we find some positive effects from increased gathering at grocery stores, negative effects of visiting retail stores and workplaces, and even small effects on visiting parks highlighting the complexities travel and migration have on the transmission of diseases.
我们研究了经济、人口和流动性相关因素在2020年对新冠病毒传播的影响。虽然学术文献中的许多模型采用线性/广义线性模型，但很少有研究纳入空间分析，而空间分析对于在疫苗推出之前理解影响疾病扩散的因素很有用。为此，我们使用了泊松广义线性模型并结合空间自回归结构。我们的分析得出了一些见解，包括在该国的一些地区，出现了呆在家里会导致疾病扩散增加这一与直觉相悖的结果。此外，我们发现杂货店聚集增加有一些积极影响，访问零售店和工作场所有负面影响，甚至去公园也有微小影响，这突出了出行和迁移对疾病传播的复杂性。
