Fill Intensity Hawkes

What problem does it solve?

This skill replaces simplistic fill-rate assumptions with a Hawkes-based model that captures self-excitation, state-dependent baselines, and queue-position effects to estimate fill intensity for market-making decisions.

Core Features & Use Cases

• State-dependent baseline intensity μ(t) driven by funding, OI changes, settlement phase, and volatility
• Trade-type dependent excitation to differentiate effects of assortments of orders
• Queue-position kernel that adapts to front-of-queue risk
• Online parameter estimation to adapt to changing market conditions
• Intensity-to-kappa conversion for GLFT integration and kappa-aware quoting
• Calibration and validation tooling (Brier score, information ratio)

Quick Start

Configure and run the Hawkes-based fill intensity model using historical trades and market state data to start generating forecasted fill probabilities.