Bayesian dynamic network actor models with application to South Korean COVID-19 patient movement data

Area of Science:

• Epidemiology
• Network Science
• Bayesian Statistics

Background:

• The COVID-19 pandemic necessitates understanding disease transmission dynamics.
• Analyzing infected individuals' movements is crucial for early pandemic insights.

Purpose of the Study:

• Introduce Bayesian dynamic network actor models for analyzing infected individuals' movements.
• Model relational event data using network statistics to understand movement patterns.

Main Methods:

• Utilized a fully probabilistic Bayesian approach.
• Employed network statistics to capture movement event structures between municipalities.
• Analyzed movement data from South Korea between January and March 2020.

Main Results:

• Quantified uncertainties in relational tendencies of movement events.
• Identified key movement patterns and directed flows of individuals.
• Provided insights into disease spread mechanisms in South Korea.

Conclusions:

• Bayesian dynamic network actor models offer a robust framework for analyzing infectious disease spread.
• Understanding early patient movement patterns is vital for pandemic control strategies.
• The study highlights the utility of network analysis in epidemiological research.