Hidden Risks and Optionalities in American Options

Battle Count: 65

Highly relevant for quantitative trading desks dealing with American-style options, currency options, and interest rate derivatives. The paper directly addresses practical problems encountered by option traders (currency rate flips, stock squeezes, funding constraints). The fugit-based heuristic provides a computationally tractable way to quantify hidden optionality that conventional systems miss. The framework is immediately applicable to: (1) pricing American options more accurately, (2) identifying model risk in existing pricing systems, (3) calibrating hedge ratios for American options under rate uncertainty, (4) stress testing option portfolios, and (5) understanding the true cost of early exercise flexibility. The practitioner episodes provide concrete trading scenarios where the framework prevents significant losses.

Moderate complexity. The core methodology involves: (1) building a CRR binomial lattice (well-understood), (2) computing the fugit distribution via forward probability propagation (novel but straightforward algorithm provided), (3) numerical integration via Gauss-Hermite quadrature (standard technique), and (4) rate model calibration (moment matching for normal/lognormal, closed-form for Hull-White). The main complexity lies in correctly implementing the stopping time distribution and handling multiple rate dynamics. The paper provides a complete algorithm (Section VII-A) that reduces implementation ambiguity. No specialized hardware or software is required beyond standard numerical computing tools.

The paper provides detailed parameter values (Tables I-VI), a complete algorithm for the fugit distribution (Section VII-A), and specifies numerical methods (binomial lattice with n=2000, Gauss-Hermite quadrature). However, several cross-references are marked with '??' indicating incomplete sections. The code is not publicly available. The methodology is described in sufficient detail for reimplementation, but some implementation specifics (e.g., exact quadrature nodes, handling of correlated rates) are left implicit. The FRED data source for interest rates is cited.