Neural Networks for Parameter Estimation of the Discretely Observed Hawkes Process
Jason J. Lambe, Feng Chen, Tom Stindl, Tsz-Kit Jeffrey Kwan · 2025
Read Neural Networks for Parameter Estimation of the Discretely Observed Hawkes Process by Jason J. Lambe, Feng Chen, Tom Stindl, Tsz-Kit Jeffrey Kwan (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-01258.txt.
Description
Jason Lambe and co-authors estimate the settings of a Hawkes process, a model for events that trigger more events, when only counts per time block are known. With exact event times hidden, the usual likelihood math becomes unworkable. Their 2025 method skips the likelihood and trains a neural network on simulated data instead.
A quick rough estimate from the counts serves as the network's input summary, needing little expert design. The trained network matches the best existing approximate methods on error, while running much faster. A quantile version gives quick uncertainty bands as well. The team apply the method to weekly counts of two infectious diseases, with a seasonal background rate.
Catalog record
| CREATOR | Jason J. Lambe, Feng Chen, Tom Stindl, Tsz-Kit Jeffrey Kwan |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 67 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-01258