11/24/2021
By Sokny Long

The UMass Lowell Francis College of Engineering, Electrical & Computer Engineering, invites you to attend a master’s thesis defense by Emi Aoki on “A Statistical Characterization of COVID-19 Infections Considering Population Density, Race and Ethnicity, and Age Groups as Covariates.”

MSE Candidate: Emi Aoki
Defense Date: Tuesday, Dec. 7, 2021
Time: 3 to 4 p.m. EST
Location: This will be a virtual defense via Zoom. Those interested in attending should contact the student, Emi_Aoki@student.uml.edu, and committee advisor, Kavitha_Chandra@uml.edu, at least 24 hours prior to the defense to request access to the meeting.

Committee Chair (Advisor): Kavitha Chandra, Associate Dean of Undergraduate Affairs, Electrical & Computer Engineering, UMass Lowell

Committee Members:

  • Charles Thompson, Professor, Electrical & Computer Engineering, UMass Lowell
  • Joshua Levy, Adjunct Professor, Electrical & Computer Engineering, UMass Lowell
  • Max Denis, Assistant Professor, Mechanical Engineering, University of the District of Columbia

Brief Abstract:

This study combines data from the US Census Bureau, Wisconsin Department of Health Services, and Wisconsin Department of Administration to analyze the state-level COVID-19 infections as a function of the city and/or county size, its total population, age, race and ethnicity demographics and job classification. The State of Wisconsin originally published only the total number of daily infections but more recently disaggregated this information by age, race, ethnicity, and percentage of health-care workers that were infected.

Recently, several studies have presented evidence that COVID-19 infections have disproportionately affected people of color. A two-phase study is carried out in this research considering aggregated infection data in the first phase and the disaggregated data in the second phase to identify statistical relationships and models that can improve our understanding on differential impact of the COVID-19.

In the first phase, a principal component analysis (PCA) of 18 features consisting of age, race and ethnicity, job classification, and population density is conducted to identify the linear combination of features that lead to a reduced set of dimensions. The magnitude and angle of the PCA score against the number of COVID-19 infection plots provide an initial classification of cities into 3 clusters that are further analyzed for their features. The reduced feature set is applied in linear and nonlinear regression models to predict the aggregate COVID infections. In the second phase, the disaggregated data that provides the daily number of infections as a function of race/ethnicity and age is analyzed using conditional probabilities to refine a prior assumption of the distribution of COVID infections among the different race, ethnicity, and age group variables.

All interested students and faculty members are invited to attend the online defense via remote access.