Calibrating Inelastic Markets to Options: The Lean Marketron and the Generalized Langevin Equation

Battle Count: 50

The paper is primarily a model calibration and identifiability study rather than a direct trading strategy paper. However, it provides a structurally identifiable framework for measuring market price of flow risk, calibrating option surfaces with a single parameter set, and understanding price impact dynamics. The reduced 9-parameter model could serve as a foundation for flow-aware trading signals, risk management in inelastic markets, and structural option pricing. The practical trading relevance is moderate as the paper explicitly states results are an 'existence proof for practical identifiability, not a trading manual or a backtested strategy.'

High complexity: requires solving a 2D PDE (Hamilton-Jacobi-Bellman) with state-dependent volatility, mixed derivatives, and nonlinear gradient-square terms. Multiple numerical approaches are discussed (Strang splitting, Cole-Hopf transformation, RBF methods, explicit FD). The JAX-based differentiable pipeline with automatic differentiation through the pricing scheme adds engineering complexity. The calibration involves bound-constrained optimization (SLSQP), profile likelihood computation, Hessian eigenspectrum analysis, and MBAM geodesic integration. The 70x50x500 grid and CFL stability constraints require careful implementation.

The paper provides detailed parameter tables, explicit PDE formulations, JAX-based implementation description, and comprehensive appendices with proofs. However, no GitHub repository or code is explicitly provided. The SPX data from January 18, 2017 is standard market data. The numerical scheme (explicit FD in JAX, 70x50x500 grid) is fully specified. Reproduction would require significant effort in implementing the 2D PDE solver and calibration pipeline.