Quantum computing for multidimensional option pricing: End-to-end pipeline

Battle Count: 69

The paper is highly relevant to quantitative finance and derivatives trading desks. It provides a practical, arbitrage-aware framework for pricing multi-asset exotics (basket, spread, worst-of options) that are actively traded. The quantum acceleration (10-100x fewer queries) could significantly reduce computational costs for real-time pricing and risk management. However, the current quantum advantage is demonstrated only via simulation, not on actual hardware, limiting immediate practical deployment. The NIG calibration methodology and copula-based joint distribution construction are directly applicable to trading desk workflows. The work is more relevant to derivatives pricing and risk management than to directional trading strategies.

High implementation complexity due to: (1) NIG Lévy process calibration requiring constrained nonlinear optimization with Tikhonov regularization; (2) quantum state preparation for loading density functions and payoff functions into quantum amplitudes; (3) QAE algorithm implementation (modified RQAE and IQAE variants); (4) cosine-series density estimation with proper truncation and endpoint handling; (5) Gaussian copula density computation in N dimensions; (6) integration of classical and quantum components in a coherent pipeline. Requires expertise in both quantitative finance (Lévy processes, arbitrage theory, copulas) and quantum computing (circuit design, amplitude estimation, state preparation). The myQLM/NEASQC software stack helps but quantum circuit construction remains non-trivial.

The paper provides detailed calibration parameters, market data sources (Euronext, Yahoo Finance), software specifications (Python 3.10, myQLM 1.12.2, NEASQC library), and hardware details (Intel Core Ultra 9 285H, 64GB RAM). However, no code repository is explicitly linked, and quantum simulation results depend on specific simulator implementations. The theoretical proofs are complete and self-contained. Market data from Euronext may require subscription access.