Fine-Grained $\epsilon$-Margin Closed-Form Stabilization of Parametric Hawkes Processes

Abstract

Hawkes Processes have undergone increasing popularity as default tools for modeling self- and mutually exciting interactions of discrete events in continuous-time event streams. A Maximum Likelihood Estimation (MLE) unconstrained optimization procedure over parametrically assumed forms of the triggering kernels of the corresponding intensity function are a widespread cost-effective modeling strategy, particularly suitable for data with few and/or short sequences. However, the MLE optimization lacks guarantees, except for strong assumptions on the parameters of the triggering kernels, and may lead to instability of the resulting parameters .In the present work, we show how a simple stabilization procedure improves the performance of the MLE optimization without these overly restrictive assumptions.This stabilized version of the MLE is shown to outperform traditional methods over sequences of several different lengths.

arXiv commentPresented as a RobustML workshop paper at ICLR 2021

References in corpus (9)

• Diffusion of Earthquake Aftershock Epicenters, Omori's Law and Generalized Continuous-Time Random Walk Models
• Non-parametric kernel estimation for symmetric Hawkes processes. Application to high frequency financial data
• Fully Neural Network based Model for General Temporal Point Processes
• Discovering Latent Network Structure in Point Process Data
• Scaling limits for Hawkes processes and application to financial statistics
• Learning Network of Multivariate Hawkes Processes: A Time Series Approach
• Tick: a Python library for statistical learning, with a particular emphasis on time-dependent modelling
• Learning Hawkes Processes from a Handful of Events
• Hawkes Process Kernel Structure Parametric Search with Renormalization Factors