Modeling the Spatial Factors of COVID-19 in New York City

Abstract

The novel coronavirus (COVID-19) is currently being regarded worldwide as a global pandemic, with New York City as one of the epicenters. There is an emerging demand for precise investigation on this disease, specifically how to slow outbreaks and reopen responsibly. This paper explores and models the spatial factors of COVID-19 in New York City through ordinary least squares regression and geographically weighted regression. Results indicate medical density, green space density, mean distance traveled, male percentage, and commuting (walking, carpooling, and public transit) could correlate to higher rates of COVID-19 positive cases. In contrast, areas with high percentages working from home and white only could correlate to lower rates of COVID-19 positive cases. Additionally, there are distinct associations in various zip code areas or clusters. Overall, this study suggests that public sanitation is critical in disease control in areas with high public transportation demand, and the effect of travel reduction is significant in delaying the outbreak. This study advises policymakers to implement unique policies, preventions measures, and reopening strategies based on localized situations considering COVID-19 outbreaks.

Keywords: Geographically weighted regression, COVID-19, spatial factors