Statistics Senior Thesis Defense by Samantha Kilcoyne ’23

Co-Occurrences of Long COVID Symptoms: An Analysis Using the Ising Model and Nodewise Logistic Regression by Samantha Kilcoyne ’23, Statistics Senior Thesis Defense, Wednesday, May 10, 1:10 – 1:50 pm, North Science Building 114, Wachenheim.

Abstract:  As the COVID-19 pandemic continues, more researchers are studying the lasting symptoms of COVID-19 in those who survive the acute phase of the infectious disease, known as Long COVID or Post-COVID Conditions (PCC).  We aim to contribute to this ongoing body of work by estimating a network of symptoms experienced by patients in the post-acute phase of COVID-19.  To estimate the network, we consider pairwise interactions of PCC through penalized logistic regression.  Thus, this thesis rigorously examines the methodology of the eLasso Method, an extension of the Ising Model and Nodewise Logistic Regression.  Finally, we analyze the network of Long COVID symptoms through community detection and various node-level network characteristics.