B1406
Title: A Bayesian approach to space- and time-indexed Markov processes, with application to the Italian premier football league
Authors: Michael Schweinberger - Pennsylvania State University (United States) [presenting]
Guanyu Hu - The University of Texas Health Science Center at Houston (United States)
Abstract: Technological advances have paved the way for collecting a wealth of data on interactions among team players in football, baseball, and other team-based sports. The resulting data involve networks of interactions within and between opposing teams and are indexed by space and time. Such space- and time-indexed network data are vital to understanding and predicting the performance of teams, because a team's performance is more than the sum of the strengths of its players. We pursue a Bayesian approach to modeling entire games of opposing sport teams as space- and time-indexed continuous-time Markov processes. We present an application to data recorded during the 2020/2021 season of the Italian premier football league (Serie A), which includes some of the best-known teams in European football.
