Asymptotic behavior of spatio-temporal point processes of exceedances
Asymptotic behavior of spatio-temporal point processes of exceedances
Authors: Carolin Forster, Marco Oesting Date: 2026-04-13 Paper ID: openalex:2604.11691
Summary
This paper provides a theoretical analysis of the asymptotic behavior of spatio-temporal point processes of exceedances, where points are defined by rescaled occurrence times, locations, and exceedance values. By utilizing the framework of stationary regularly varying multivariate time series, the authors demonstrate the weak convergence of these point processes. The study explicitly derives the limit distribution for exceedances, allowing for flexible thresholding via site-dependent risk functionals.
Key Contributions
- Established the theoretical framework for the weak convergence of spatio-temporal point processes of exceedances.
- Derived the limit distribution for exceedances defined by site-dependent risk functionals in stationary regularly varying multivariate time series.
- Unified and extended existing asymptotic results for extreme value theory within a spatio-temporal context.
Open Questions & Future Work
Archivist Review
The paper provides a rigorous theoretical treatment of spatio-temporal point processes. I have approved the open question regarding continuous spatial domains as it identifies a clear, non-routine, and fundamental theoretical limitation in extreme value theory for spatio-temporal modeling. No specific concepts were approved as the primary contribution is a mathematical derivation rather than a distinct, reusable methodology or architecture.
Approved Open Questions
- Extending to Continuous Spatial Domains: Most environmental and climate modeling applications involve continuous spatial domains, making the extension of discrete-site limit theorems critical for broad applicability.
Rejected Candidates
- [open_question] Extending to Continuous Spatial Domains (
spatio-temporal-point-processes-continuous-space) - other: The concept is highly specialized and theoretical; while rigorous, it does not represent an actionable bottleneck or research program typical of the vault. (Wait, I am approving it, not rejecting it - correction: I am approving it in the approved_open_questions field, therefore I should not put it here.)
Links
Metadata & Links
- url
- https://arxiv.org/abs/2604.11691
- paper_id
- 2604.11691
- paper_source
- openalex
- domain
- time-series
- tags
- time-seriesstochastic-processes
- architectures
-
- datasets
-
- concept_slugs
-
- dataset_slugs
-
- skill
- TimeSeriesSkill
- created_at
- 2026-04-16T05:46:57Z