Paper: arXiv 2307.04510

Authors: Christian Yeo

Abstract

Linear regression, firstly introduced for the pricing of American-style options, has since been expanded to include swing options pricing. Swing options price may be viewed as the solution to a Backward Dynamic Programming Principle, which involves a conditional expectation known as the continuation value. The approximation of the continuation value using linear regression involves two levels of approximation. First, the continuation value is replaced by an orthogonal projection over a subspace spanned by a finite set of $m$ squared-integrable functions yielding a first approximation $V^m$ of the swing value function. In this paper, we prove that, with well-chosen regression functions, $V^m$ converges to the swing actual price $V$ as $m \to + \infty$. A similar result is proved when classic regression functions are replaced by neural networks. For both methods (linear regression and neural networks), we analyze the second level of approximation involving practical computation of the swing price using Monte Carlo simulations and yielding an approximation $V^{m, N}$ (where $N$ denotes the Monte Carlo sample size). Especially, we prove that $V^{m, N} \to V^m$ as $N \to + \infty$ for both methods and using a Hilbert basis assumption in the linear regression. Besides, a convergence rate of order $\mathcal{O}\big(\frac{1}{\sqrt{N}} \big)$ is proved in the linear regression case.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — theoretically deep, empirically untested

Why this score: The paper provides rigorous mathematical proofs establishing convergence rates and theoretical foundations for pricing swing options using linear regression and neural networks. However, it lacks empirical validation, backtesting on market data, or code implementations to assess real-world performance.

Research Flowchart

  flowchart TD
    A["Research Goal: Price Swing Options via Regression & Neural Networks"] --> B["Data Inputs: Underlying Asset Paths (Monte Carlo N) & Payoff Functions"]
    B --> C["Level 1 Approximation: Project Continuation Value onto Basis (m Functions / NN)"]
    C --> D["Level 2 Approximation: Compute Price V<sup>m,N</sup> using N Monte Carlo Simulations"]
    D --> E1["Outcome 1: Theoretical Convergence<br>(V<sup>m</sup> → V as m → ∞ & V<sup>m,N</sup> → V<sup>m</sup> as N → ∞)"]
    D --> E2["Outcome 2: Convergence Rate<br>(Order O(1/√N) for Linear Regression)"]