SARS-COV-2 DISSEMINATION USING UNITED STATES COUNTY COMMUTING DATA
Autor: | Urrutia, Patrick M. |
---|---|
Přispěvatelé: | Yoshida, Ruriko, Vogiatzis, Chrysafis, University of Illinois, Royset, Johannes O., Operations Research (OR) |
Rok vydání: | 2021 |
Předmět: |
principal component analysis
commute county naïve virus naïve model ARIMA supervised learning epidemic mean absolute percentage error GNAR commuting MASE mean absolute scaled error travel COVID epidemiologist transportation PCA algorithm generalized network autoregressive SARS-CoV-2 pandemic MAPE COVID-19 naive model Holt-Winters naive network epidemiology time series Python |
Popis: | The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has impacted the world for the past year and a half, and it has been spreading to all corners of the Earth. Analyzing the dissemination of SARS-CoV-2 can allow leaders of certain areas to preemptively enact measures that could prevent the virus from spreading further. By analyzing commuting data between counties in the United States, one can create a predictive model that will allow interdiction of routes with high traffic between areas to stop the spread of the virus. At the county level, leaders can use this information to provide extra precautions, medical equipment, and testing in their area of jurisdiction. We solve this problem by obtaining data about coronavirus-19 (COVID-19) cases and deaths from the Center for Disease Control and Prevention and county commuting data from the United States Census Bureau. Then we propose to apply the generalized network autoregressive (GNAR) time series model for analyzing this network over time series data. This by-county predictive approach is broken down by state, in order to reflect more localized trends. This thesis combines time series analysis and network science to model COVID-19 cases and deaths by state. Ensign, United States Navy Approved for public release. Distribution is unlimited. |
Databáze: | OpenAIRE |
Externí odkaz: |