Fill Intensity Hawkes

Model trade fill intensity for high-frequency market making using a Hawkes-based approach that captures self-excitation, state-dependent baselines, and queue-position effects to estimate fill probabilities from historical trades and market state data.

A state-aware Hawkes process improves fill probability prediction by conditioning baseline intensity on funding, open interest changes, settlement phase, and volatility, while applying trade-type dependent excitation to differentiate order effects.

Yes, you can use online parameter estimation for Hawkes fill intensity calibration to adapt to changing market conditions, alongside offline MLE fitting for initial baseline, excitation, and kernel component estimation.

Convert fill intensity to kappa for GLFT framework integration using the provided conversion method, enabling kappa-aware quoting and state-dependent market-making decisions based on forecasted fill probabilities.

Calibrating a Hawkes fill intensity model requires historical fills and trades data, a current market state representation including funding and OI, and implementations for baseline, excitation, and kernel components.

Validation metrics for fill intensity prediction models include Brier score and information ratio, provided as calibration and validation tooling to assess the accuracy of forecasted fill probabilities.