Scalable Durational Event Models: Application to Physical and Digital Interactions
Cornelius Fritz, Riccardo Rastelli, Michael Fop, Alberto Caimo
[stat.ME,stat.CO]
Durable interactions are ubiquitous in social network analysis and are increasingly observed with precise time stamps. Phone and video calls, for example, are events to which a specific duration can be assigned. We refer to this type of data encoding the start and end times of interactions as “durational event data”. Recent advances in data collection have enabled the observation of such data over extended periods of time and between large populations of actors. Building on Relational Event Models, we propose the “Durational Event Model” as a framework for studying durational events by separately modeling event incidence and duration. To accommodate large-scale applications, we introduce a fast, memory-efficient, and exact block-coordinate ascent algorithm. Theoretical and numerical results demonstrate several advantages of this approach over traditional Newton-Raphson-based methods. We apply the model to physical and digital interactions among college students in Copenhagen. Our empirical findings reveal that past interactions are the main drivers of physical interactions, whereas digital interactions are more strongly influenced by friendship ties and prior dyadic contact.