Paper: arXiv 2611.15564

Authors: Yuming Ma, Syoiti Ninomiya

Abstract

High-order recombination keeps weak approximation trees of diffusions small by a moment-preserving reduction of the measure on each patch of nodes. The weighted Litterer–Lyons (WLL) test of Ninomiya and Shinozaki sets the patch scale using a scalar \(λ>0\) chosen beforehand, whose appropriate value depends on coordinate units. First, the anisotropically weighted Litterer–Lyons (AWLL) rule selects the scale during recombination within a bounded factor of the measured increment amplitude. On a fixed set of nodes, expanding coordinate units leaves the AWLL patch-count bound unchanged but raises the WLL lower bound as a power until saturation. In Heston and shifted Stochastic-Alpha-Beta-Rho (sSABR) systems with initial price or forward rate 100, AWLL reduces the terminal supports of WLL with \(λ=1\) by factors of 137 and 1488 at comparable errors, but enlarges them at smaller Heston initial prices. Second, \(λ\) is extended to a positive definite matrix with a joint error bound for prices and sensitivities, a bound on the gain of its shape, and patch-count bounds. Third, the elimination stage of batched basis exchange requires \(O(J)\) arithmetic operations for \(J\) nodes at fixed moment rank and panel width, against \(Θ(J^{3})\) for the direct kernel update of Ninomiya and Shinozaki.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper presents a highly mathematical and theoretically dense approach to weak approximation of SDEs, demonstrating significant novelty in its adaptive scale selection and computational efficiency. While it includes empirical results on Heston and sSABR models, the focus is more on theoretical improvements and algorithmic complexity rather than extensive backtesting or data validation. The methodology is detailed, but explicit code or data availability is not mentioned in the provided excerpt.

Research Flowchart

  flowchart TD
    A[Research Goal: Improve Weak Approximation of SDEs for Derivative Pricing] --> B{Problem: WLL Scale Selection Depends on User-Defined Scalar \(λ\) & Coordinate Units};
    B --> C[Methodology 1: Anisotropically Weighted Litterer-Lyons (AWLL) Rule];
    C --> D{Computational Process: Adaptive Scale Selection Based on Increment Amplitude};
    D --> E[Data/Inputs: Heston & sSABR Systems (Initial Price/Forward Rate 100)];
    E --> F{Key Finding 1: AWLL Reduces Terminal Supports of WLL (λ=1) by 137x (Heston) & 1488x (sSABR) at Comparable Errors};
    F --> G{Key Finding 2: AWLL Robust to Coordinate Unit Changes; WLL Lower Bound Raises as Power Until Saturation};
    G --> H[Methodology 2: \(λ\) Extended to Positive Definite Matrix for Joint Error/Gain/Patch-Count Bounds];
    H --> I[Methodology 3: \(O(J)\) Arithmetic for Batched Basis Exchange (Elimination Stage)];
    I --> J{Overall Outcome: More Efficient, Robust, and Adaptable High-Order Recombination for Weak Approximation Trees};