First-passage horizons in horizontal visibility graphs: a rank-invariant estimator of path roughness for rough volatility models

Battle Count: 78

Directly relevant to volatility modeling and rough volatility literature. Provides a rank-invariant, parameter-free diagnostic for distinguishing rough Bergomi from classical Heston, GARCH, and FIGARCH models. The VIX application demonstrates practical utility for volatility regime detection. However, it is a diagnostic tool rather than a direct trading signal generator, and the FIGARCH separation caveat limits its use for model selection in some regimes.

The core L+(t) computation is O(T) via monotone stack (next-greater-or-equal problem). The Hill-MLE with CSN threshold selection is standard. Block bootstrap adds moderate complexity. Simulation of fBm (Davies-Harte) and RL-fBm (Bennedsen hybrid) requires careful implementation. Overall moderate complexity with well-documented algorithms.

The paper provides detailed simulation parameters (N=10,000, T=2^16, Davies-Harte algorithm, Bennedsen hybrid scheme), explicit estimator formulas, threshold selection protocol, and uses publicly available FRED VIX data. However, no GitHub repository is mentioned. All simulation details are specified in Appendix C.