Spatiotemporal Hawkes processes with a graphon-induced connectivity structure
Justin Baars, Roger J.A. Laeven, Michel Mandjes

TL;DR
This paper introduces a novel spatiotemporal Hawkes process influenced by a graphon structure, establishing its theoretical properties, limits, and convergence behaviors, thus generalizing existing Hawkes models to infinite-dimensional settings.
Contribution
It develops a new graphon-based Hawkes process model, proves its existence, stability, and limit behaviors, and connects finite-dimensional Hawkes processes with the infinite-dimensional limit.
Findings
The process is well-defined, unique, and stable.
Finite-dimensional Hawkes processes converge to the graphon Hawkes process as dimension increases.
Limit behaviors include FLLN, FCLT, and divergence in unstable regimes.
Abstract
We introduce a spatiotemporal self-exciting point process , boundedly finite both over time and space , with excitation structure determined by a graphon on . This graphon Hawkes process generalizes both the multivariate Hawkes process and the Hawkes process on a countable network, and despite being infinite-dimensional, it is surprisingly tractable. After proving existence, uniqueness and stability results, we show, both in the annealed and in the quenched case, that for compact, Euclidean , any graphon Hawkes process can be obtained as the suitable limit of -dimensional Hawkes processes , as . Furthermore, in the stable regime, we establish an FLLN and an FCLT for our infinite-dimensional process on compact , while in the unstable regime we prove…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPoint processes and geometric inequalities · Diffusion and Search Dynamics · Biomedical Research and Pathophysiology
