Honors Projects

Abstract

Some might associate the term ‘public health’ with the pandemic that occurred in 2020. COVID-19 spread like most have never seen in their lifetime. It is useful to look at the effectiveness of the travel re- strictions in mitigating the spread of the global pandemic. Using linear regression and network regression, we obtain parameter estimates to determine the relation of predictors, such as network effect, percentage of urban population and GDP, on the COVID-19 incidence rate for the months January to April of 2020. Linear regression does not ac- count for the correlation structure of the data. Network regression, on the other hand, performs this task effectively, following a community detection algorithm such as the infomap method, and calculates pa- rameter estimates using the communities, or clusters, within the data. Through simulations, the consistency of both linear regression and network regression estimators is evaluated. We compare the network regression estimates to the linear regression estimates. For further re- search, we determine if the network effect is significant. The results of linear regression estimates and the network regression estimates differed, though still providing similar conclusions. More emphasis is placed on the estimates obtained through network regression given how it accounts for the networks within the data. Considering results from the estimates, it does not appear that the travel restrictions set around the globe assisted in minimizing the spread of COVID-19.

Department

Mathematics and Statistics

Major

Mathematics

First Advisor

Riddhi Ghosh

First Advisor Department

Mathematics and Statistics

Second Advisor

Bradley Fevrier

Second Advisor Department

Public and Allied Health

Publication Date

Fall 12-4-2023

Download

Share

COinS