Dynamic slippage control and rejection feedback in spot FX market making

Battle Count: 91

Highly relevant to FX market making desks and algorithmic trading. The paper directly addresses practical problems faced by OTC FX dealers: latency risk management, trade rejection optimization, adverse selection mitigation, and fair execution protocol design. The adiabatic-quadratic approximation provides a tractable tool for real-time policy design. The rejection feedback mechanism models realistic client-dealer dynamics. The framework bridges academic market-making theory with industry practice (last look, price validation, GFXC code compliance).

The full HJB solution requires numerical PDE methods with multi-dimensional state space (inventory, rejection score). The adiabatic-quadratic approximation significantly reduces complexity to solving a Riccati ODE and performing 1D numerical maximization for tolerance parameters. The FOC for optimal quotes requires root-finding. Monte Carlo simulation for validation adds computational cost. Practical implementation requires careful calibration of adverse selection parameters, latency distributions, and reputation dynamics.

The paper provides detailed mathematical formulations, parameter sets for numerical examples, and describes the adiabatic-quadratic approximation procedure. However, no code or data repository is provided. The Monte Carlo simulation methodology is described but not fully reproducible without implementation. Parameters are specified for a liquid currency pair scenario.