Paper: arXiv 2405.00701

Authors: Rihito Sakurai, Haruto Takahashi, Koichi Miyamoto

Abstract

A long-standing issue in mathematical finance is the speed-up of option pricing, especially for multi-asset options. A recent study has proposed to use tensor train learning algorithms to speed up Fourier transform (FT)-based option pricing, utilizing the ability of tensor trains to compress high-dimensional tensors. Another usage of the tensor train is to compress functions, including their parameter dependence. Here, we propose a pricing method, where, by a tensor train learning algorithm, we build tensor trains that approximate functions appearing in FT-based option pricing with their parameter dependence and efficiently calculate the option price for the varying input parameters. As a benchmark test, we run the proposed method to price a multi-asset option for the various values of volatilities and present asset prices. We show that, in the tested cases involving up to 11 assets, the proposed method outperforms Monte Carlo-based option pricing with \(10^6\) paths in terms of computational complexity while keeping better accuracy.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces advanced tensor train decomposition and learning algorithms (TCI) with dense mathematical formulation, earning a high math score; it includes concrete computational benchmarks on multi-asset options with 11 assets and claims performance gains over Monte Carlo, which provides empirical rigor but lacks real-world data or code disclosure.

Research Flowchart

  flowchart TD
  A["Research Goal: Speed-up<br>Multi-asset Option Pricing"] --> B["Methodology: Tensor Train Learning<br>for Fourier-based Pricing"]
  B --> C{"Input Data: Multi-asset Option<br>Parameter Sets"}
  C --> D["Compute Tensor Train<br>Approximations"]
  D --> E["Efficiently Calculate<br>Option Prices"]
  E --> F{"Benchmarking"}
  F -->|"Monte Carlo (10^6 paths)"| G["Standard Method"]
  F -->|"Proposed Method"| H["Optimized Method"]
  G --> I["Key Findings"]
  H --> I
  I --> J["Proposed method outperforms<br>Monte Carlo in computational complexity<br>for up to 11 assets"]
  I --> K["Maintains higher accuracy<br>at lower cost"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a tensor‑train learning based pricing method that builds tensor trains to approximate functions in Fourier‑transform based option pricing with parameter dependence, enabling fast evaluation for varying inputs.

Source sentence
Here, we propose a pricing method, where, by a tensor train learning algorithm, we build tensor trains that approximate functions appearing in FT-based option pricing with their parameter dependence and efficiently calculate the option price for the varying input parameters.

Assumes 4

  • Underlying asset dynamics follow the Black‑Scholes SDE with constant risk‑free rate and constant volatilities. model
    Source sentence
    ... where r ∈R and σ1, · · · , σd > 0 are constant parameters called the risk‑free interest rate and the volatilities, respectively.
  • Brownian motions driving assets have a constant correlation matrix ρij. model
    Source sentence
    W1(t), · · · , Wd(t) are the Brownian motions with constant correlation matrix ρij, namely dWm(t)dWn(t) = ρmndt,
  • Option considered is European with payoff max(c(∏S_i(T)−K),0). other
    Source sentence
    …(T) −K), 0}, where c = 1 and −1 for a call and put option, respectively.
  • Parameter grid for volatilities and initial stock prices is discretized on integer‑labeled points. data
    Source sentence
    we assume that, in reality, the grid points in Rd is labeled by integers and the indexes of the tensor denotes the integers.

Data

  • Multi‑asset option benchmark (volatility and stock‑price variations) synthetic unknown
    up to 11 assets
    Source sentence
    As a benchmark test, we run the proposed method to price a multi‑asset option for the various values of volatilities and present asset prices.
  • Monte Carlo reference (10^6 paths) synthetic unknown
    up to 11 assets
    Source sentence
    the proposed method outperforms Monte Carlo‑based option pricing with $10^6$ paths in terms of computational complexity while keeping

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The TT‑based method outperforms Monte Carlo with 10^6 paths in computational complexity for up to 11 assets while maintaining comparable or better accuracy. performance
    Source sentence
    we show that, in the tested cases involving up to 11 assets, the proposed method outperforms Monte Carlo‑based option pricing with $10^6$ paths in terms of computational complexity while keeping
  • In all tested cases the error of the TT‑based result is smaller than that of the MC‑based result. pricing-calibration
    Source sentence
    the error in the TT‑based result is smaller than that in the MC‑based result in all the tested cases.
  • TT‑based option pricing demonstrates superiority over Monte Carlo in both computational complexity and runtime for d=5 and d=10 after SVD compression. performance
    Source sentence
    TT‑based option pricing demonstrates superiority over the Monte Carlo method in terms of computational complexity and time.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 24 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 2 papers in the archive (2 by independent authors): Full grid solution for multi-asset options pricing with … (uses its method or data); Trading Devil Final: Backdoor attack via Stock market and … (uses its method or data)

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (56)

  1. A highly efficient tensor network algorithm for multi-asset Fourier options pricing (2022) — link
  2. $O(d\log N)$ -Quantics Approximation of N-d Tensors in High-Dimensional Numerical Modeling (2011) — link
  3. Tensor Numerical Methods in Scientific Computing (2018) — link
  4. The ITensor Software Library for Tensor Network Calculations (2022) — link
  5. tf-quant-finance (2023)
  6. Options futures and other derivatives (2003)
  7. Stochastic calculus for finance I & II (2004)
  8. Quantum computing for finance (2023) — link
and 48 more
  1. Quantum speedup of Monte Carlo methods (2015) — link
  2. Option valuation using the fast Fourier transform (1999)
  3. A simple option formula for general jump-diffusion and other exponential L 'e vy processes (2001)
  4. A practical introduction to tensor networks: Matrix product states and projected entangled pair states (2014) — link
  5. Developments in the Tensor Network — from Statistical Mechanics to Quantum Entanglement (2022) — link
  6. Supervised Learning with Tensor Networks (2016) — link
  7. TTOpt: A Maximum Volume Quantized Tensor Train-based Optimization and its Application to Reinforcement Learning (2022) — link
  8. Exponential machines (2016) — link
  9. Tensor-Train Decomposition (2011) — link
  10. TT-cross approximation for multidimensional arrays (2010) — link
  11. Analysis of Fourier Transform Valuation Formulas and Applications (2010) — link
  12. Monte Carlo methods in financial engineering (2004)
  13. Tensor-Train Decomposition (2011) — link
  14. Efficient classical simulation of the approximate quantum Fourier transform (2007) — link
  15. Multiscale Space-Time Ansatz for Correlation Functions of Quantum Systems Based on Quantics Tensor Trains (2023) — link
  16. Parallel cross interpolation for high-precision calculation of high-dimensional integrals (2020) — link
  17. TT-cross approximation for multidimensional arrays (2010) — link
  18. Learning Feynman Diagrams with Tensor Trains (2022) — link
  19. Quantum Fourier Transform Has Small Entanglement (2023) — link
  20. Superfast Fourier Transform Using QTT Approximation (2012) — link
  21. Quantics Tensor Cross Interpolation for High-Resolution Parsimonious Representations of Multivariate Functions (2024) — link
  22. The density-matrix renormalization group in the age of matrix product states (2011) — link
  23. Quantum-Inspired Tensor Neural Networks for Partial Differential Equations (2022) — link
  24. Low-Rank Tensor Approximation for Chebyshev Interpolation in Parametric Option Pricing (2020) — link
  25. Solving high-dimensional partial differential equations using deep learning (2018) — link
  26. Pricing High-Dimensional Bermudan Options with Hierarchical Tensor Formats (2023) — link
  27. Quantum annealing in the transverse Ising model (1998) — link
  28. A quantum-inspired approach to exploit turbulence structures (2022) — link
  29. Quantum-inspired method for solving the Vlasov-Poisson equations (2022) — link
  30. Multiscale Space-Time Ansatz for Correlation Functions of Quantum Systems Based on Quantics Tensor Trains (2023) — link
  31. Compactness of quantics tensor train representations of local imaginary-time propagators (2024) — link
  32. Linear image coding for regression and classification using the tensor-rank principle (2001) — link
  33. Multilinear Analysis of Image Ensembles: TensorFaces (2002)
  34. Tensor-train approximation of the chemical master equation and its application for parameter inference (2021) — link
  35. Learning tensor networks with tensor cross interpolation: new algorithms and libraries (2024) — link
  36. Numerical solution of the incompressible Navier-Stokes equations for chemical mixers via quantum-inspired Tensor Train Finite Element Method (2023) — link
  37. Hierarchical Tensor Approximation of Output Quantities of Parameter-Dependent PDEs (2015) — link
  38. Pricing with a smile (1994)
  39. A closed-form solution for options with stochastic volatility with applications to bond and currency options (1993) — link
  40. Managing Smile Risk (2002)
  41. Optimal damping with a hierarchical adaptive quadrature for efficient Fourier pricing of multi-asset options in Lévy models (2023) — link
  42. Quasi-Monte Carlo for Efficient Fourier Pricing of Multi-Asset Options — → 2403.02832
  43. The Variance Gamma (V.G.) Model for Share Market Returns (1990) — link
  44. Processes of normal inverse Gaussian type (1997) — link
  45. Finite difference methods in financial engineering: a partial differential equation approach (2013)
  46. Implementing Derivative Models (1998)
  47. Smoothing the payoff for efficient computation of Basket option prices (2018) — link
  48. Preintegration via Active Subspace (2023) — link

1 of 56 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.