Mathematical Biology Seminar - Christina Wang
- When
- Thursday, November 5, 202611:10 AM – 12:10 PM MSTAdd to calendarGoogle Calendar
- Where
- Engineering Center ECCR 2571111 Engineering Drive, Boulder, CO 80309Open in Maps
Reminders arrive as notifications from the NextToU iPhone app. Get the app
About
Christina Wang, Department of Applied Mathematics, University of Colorado Boulder
Modeling the Temporal Dynamics of Protest with Self-Exciting Hawkes-Type Count Processes
Protest activity in the United States often manifests as waves shaped by both endogenous self-reinforcing dynamics and external events. While researchers have studied the temporal clustering of specific social movements, large-scale protest-event data allow systematic examination of these dynamics. This work studies the temporal structure of U.S. protest activity using daily event counts from the Crowd Counting Consortium (CCC) from 2017 to 2026. We develop a discrete-time Hawkes-type model that represents protest activity as an over-dispersed self-exciting count process while also accounting for external events, temporal seasonality, and other effects. A stepwise model comparison shows that each added component improves model fit, suggesting that these features capture distinct aspects of the protest-generating process. The fitted models also show that protest dynamics differ substantially before and after the George Floyd protest wave. The result suggests that self-exciting count models provide a useful framework for separating recurring rhythms, endogenous clustering, and external shocks in large-scale protest data.