Description: This study uses a computing platform to analyze and estimate the correlation between human mobility and COVID-19 spreading. The infection data are collected from various county, state and national sources and mobile device location data are procured from multiple third-party data providers. To capture the time-varying relationship between the number of infection and mobility inflow, the authors developed a simultaneous equations model with time-varying coefficients. Mobility data can be found in the University of Maryland COVID-19 Impact Analysis Platform. The analysis platform displays the data in a user-friendly interface utilizing a map of the United States and charts that show, under Mobility and Social Distancing, the % of in and out-of-county trips, % of in and out- of-state trips, and others. The information on the map can also be adjusted to show the mobility per state. Codebase, data of infection cases and computed metrics are shared on GitHub. The statistic modeling including data processing, prediction and visualization is written in Python and R.
Timeframe: 2020
Geographic Coverage: United States

Subject Domain

Keywords

UMB Data Catalog

Mobile device data reveal the dynamics in a positive relationship between human mobility and COVID-19 infections
UMB Dataset