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

TL;DR

This paper introduces a fine-grained stabilization method for parametric Hawkes Processes that enhances maximum likelihood estimation, ensuring stability and improved performance across various sequence lengths.

Contribution

It proposes a novel stabilization technique for MLE in Hawkes Processes, addressing instability issues without restrictive assumptions.

Findings

Stabilization improves MLE performance.

Method outperforms traditional approaches.

Effective across different sequence lengths.

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.