pith. sign in

arxiv: 2003.14184 · v76 · pith:7UHKYAZBnew · submitted 2020-03-28 · 🧮 math.PR

Strong Approximation of Iterated Ito and Stratonovich Stochastic Integrals Based on Generalized Multiple Fourier Series. Application to Numerical Solution of Ito SDEs and Semilinear SPDEs

Dmitriy F. Kuznetsov This is my paper

Pith reviewed 2026-05-24 14:18 UTC · model grok-4.3

classification 🧮 math.PR
keywords iterated stochastic integralsIto integralsStratonovich integralsmultiple Fourier seriesstrong approximationstochastic differential equationsnumerical methodsSPDEs
0
0 comments X

The pith

Generalized multiple Fourier series enable strong approximation of iterated Ito stochastic integrals for arbitrary multiplicity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a method to expand and approximate iterated Ito and Stratonovich stochastic integrals using generalized multiple Fourier series that converge in the L2 norm over the time interval raised to the power k. This approach succeeds for Ito integrals of any multiplicity k without prior restrictions on the weights. The approximations support accurate numerical solutions to Ito stochastic differential equations and semilinear stochastic partial differential equations by providing the necessary terms in Taylor expansions. A reader would care because these integrals appear in high-order numerical schemes whose error depends directly on how well the integrals are approximated.

Core claim

The central discovery is that the generalized multiple Fourier series of the integrand, converging in the L2([t, T]^k) norm, can be integrated term by term to yield a strong approximation to the iterated Ito stochastic integral of multiplicity k for every natural number k. The same technique is adapted to Stratonovich integrals under conditions on the weight functions and the choice of complete orthonormal system, with explicit mean-square error formulas derived for the approximations.

What carries the argument

The generalized multiple Fourier series expansion in L2([t,T]^k) of the product of the weight function and the indicator functions for the integration domain, which produces an explicit series representation of the iterated integral.

If this is right

  • Exact and approximate expressions for the mean-square error of approximation of Ito ISIs of multiplicity k for any natural number k.
  • Adaptation of the expansions to Stratonovich ISIs of multiplicities 1 to 8 with continuously differentiable weights and Legendre or trigonometric bases.
  • Adaptation to Stratonovich ISIs of multiplicities 1 to 6 with continuous weights and arbitrary complete orthonormal systems.
  • Application of the expansions to Taylor-Ito and Taylor-Stratonovich series for numerical integration of Ito SDEs and semilinear SPDEs.
  • Extension of the approximations to iterated integrals with respect to the Q-Wiener process.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Higher-multiplicity terms in stochastic Taylor expansions become feasible for simulation once the series truncation error is controlled.
  • The same orthogonal expansion technique may transfer to other iterated integrals arising in non-Markovian or jump-driven models.
  • Implementation cost for large k can be reduced by exploiting fast evaluation of the chosen orthonormal system.
  • Error bounds derived from the L2 remainder could be combined with existing strong convergence theory for SDE solvers to obtain overall scheme orders.

Load-bearing premise

The generalized multiple Fourier series of the integrand converges in L2 norm to the iterated integral itself for arbitrary multiplicity k, without additional restrictions on the weight functions beyond those stated for the Stratonovich case.

What would settle it

A concrete weight function and multiplicity k greater than 8 where the L2 remainder after any finite truncation of the series does not approach zero in mean square when compared against a direct Monte Carlo estimate of the iterated integral.

read the original abstract

The book is devoted to the strong approximation of iterated stochastic integrals (ISIs) in the context of numerical integration of Ito SDEs and non-commutative semilinear SPDEs with nonlinear multiplicative trace class noise. The monograph opens up a new direction in researching of ISIs. For the first time we successfully use the generalized multiple Fourier series converging in the sense of norm in $L_2([t, T]^k)$ for the expansion and strong approximation of Ito ISIs of multiplicity $k,$ $k\in{\bf N}$ (Chapter 1). This result has been adapted for Stratonovich ISIs of multiplicities 1 to 8 (the case of continuously differentiable weight functions and a CONS of Legendre polynomials or trigonometric functions in $L_2([t, T])$) and for Stratonovich ISIs of multiplicities 1 to 6 (the case of continuous weight functions and an arbitrary CONS in $L_2([t, T])$) (Chapter 2), as well as for some other types of iterated stochastic integrals (Chapter 1). Recently (in 2024), the mentioned adaptation has also been carried out for Stratonovich ISIs of multiplicity $k$ $(k\in{\bf N})$ for the case of an arbitrary CONS in $L_2([t, T])$ (Theorems 2.59, 2.61) but under one additional condition. We derived the exact and approximate expressions for the mean-square error of approximation of Ito ISIs of multiplicity $k$, $k\in{\bf N}$ (Chapter 1). We provided a significant practical material (Chapter 5) devoted to the expansions of specific Ito and Stratonovich ISIs of multiplicities 1 to 6 from the Taylor-Ito and Taylor-Stratonovich expansions (Chapter 4) using the CONS of Legendre polynomials and the CONS of trigonometric functions. The methods formulated in this book have been compared with some existing methods (Chapter 6). The results of Chapter 1 were applied (Chapter 7) to the approximation of ISIs of multiplicity $k,$ $k\in{\bf N}$ with respect to the $Q$-Wiener process.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript develops strong approximations for iterated Ito and Stratonovich stochastic integrals (ISIs) of arbitrary multiplicity k using generalized multiple Fourier series that converge in the L2([t,T]^k) norm. Chapter 1 claims this yields expansions and mean-square error formulas for Ito ISIs without additional restrictions on the weight functions; Chapter 2 adapts the approach to Stratonovich ISIs up to multiplicity 8 (continuously differentiable weights, specific CONS) or 6 (continuous weights, arbitrary CONS), with a 2024 extension to arbitrary k under one extra condition (Theorems 2.59, 2.61). Explicit expansions for multiplicities 1–6, comparisons with existing methods, and applications to Taylor-Ito/Stratonovich expansions and Q-Wiener process approximations are provided in later chapters.

Significance. If the central convergence claims hold, the work supplies a systematic Fourier-based framework for approximating ISIs that appears to be the first to treat arbitrary multiplicity for the Ito case, together with exact mean-square error expressions and concrete expansions usable in SDE/SPDE numerics. The practical material (Chapter 5) and method comparisons (Chapter 6) add concrete value for implementation.

major comments (2)
  1. [Chapter 1] Chapter 1: The claim that generalized multiple Fourier series yield strong approximations of Ito ISIs for arbitrary k “without additional restrictions” on the weight functions must be reconciled with the extra condition required for the Stratonovich case in Theorems 2.59 and 2.61. The manuscript should explicitly verify that term-by-term integration against the Wiener process preserves L2 convergence for Ito integrals under only the stated hypotheses; if the same orthogonality or regularity property is implicitly used, the “without additional restrictions” statement is not supported.
  2. [Chapter 1] Chapter 1 (error formulas): The exact and approximate mean-square error expressions for Ito ISIs of multiplicity k must be shown to remain valid when the generalized Fourier series is integrated term-by-term; any hidden regularity assumption on the weight functions that is needed for the interchange must be stated, otherwise the error bounds are not guaranteed for arbitrary CONS.
minor comments (2)
  1. [Abstract, Chapter 2] The abstract and Chapter 2 should clarify the precise additional condition in Theorems 2.59 and 2.61 so that readers can immediately compare it with the Ito hypotheses in Chapter 1.
  2. Notation for the generalized multiple Fourier coefficients and the associated multiple integrals should be introduced once with a single consistent definition rather than re-derived in each chapter.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments. We address the major comments point by point below.

read point-by-point responses

  1. Referee: [Chapter 1] Chapter 1: The claim that generalized multiple Fourier series yield strong approximations of Ito ISIs for arbitrary k “without additional restrictions” on the weight functions must be reconciled with the extra condition required for the Stratonovich case in Theorems 2.59 and 2.61. The manuscript should explicitly verify that term-by-term integration against the Wiener process preserves L2 convergence for Ito integrals under only the stated hypotheses; if the same orthogonality or regularity property is implicitly used, the “without additional restrictions” statement is not supported.

    Authors: The Ito integral is an L2-isometry, so L2 convergence of the generalized multiple Fourier series on [t,T]^k directly transfers to the integrated process without extra regularity on the weights. The Stratonovich case requires a different limiting procedure, which is why Theorems 2.59 and 2.61 impose an additional condition for arbitrary k. We will add an explicit remark and short verification in Chapter 1 that isolates this distinction and confirms preservation of L2 convergence under the stated hypotheses for the Ito integrals. revision: yes

  2. Referee: [Chapter 1] Chapter 1 (error formulas): The exact and approximate mean-square error expressions for Ito ISIs of multiplicity k must be shown to remain valid when the generalized Fourier series is integrated term-by-term; any hidden regularity assumption on the weight functions that is needed for the interchange must be stated, otherwise the error bounds are not guaranteed for arbitrary CONS.

    Authors: The mean-square error formulas are obtained by applying the Ito isometry to the L2 remainder of the Fourier series; no further regularity on the weights is used beyond the L2([t,T]^k) convergence of the series itself. We will insert a short paragraph in Chapter 1 that recalls the isometry and justifies the term-by-term integration for arbitrary CONS, thereby confirming that the error expressions hold as stated. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected; derivation rests on standard L2 Fourier theory.

full rationale

The paper applies generalized multiple Fourier series (converging in L2([t,T]^k)) to expand Ito iterated stochastic integrals, deriving mean-square error expressions from this expansion. No quoted step shows a quantity defined in terms of itself, a fitted parameter renamed as a prediction, or a load-bearing claim justified solely by self-citation whose content reduces to the target result. The Stratonovich adaptation under an additional condition is explicitly distinguished from the Ito case in Chapter 1, and the central claims remain independent of any self-referential fitting or imported uniqueness theorems. The approach is self-contained against external Fourier analysis benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the L2 convergence of the generalized multiple Fourier series for the integrands of iterated stochastic integrals. No free parameters, invented entities, or non-standard axioms are mentioned in the abstract.

axioms (1)
  • domain assumption Generalized multiple Fourier series converge in L2([t,T]^k) to the target iterated integral for any complete orthonormal system in L2([t,T]).
    Invoked in Chapter 1 as the basis for the expansion of Ito ISIs of multiplicity k.

pith-pipeline@v0.9.0 · 5963 in / 1466 out tokens · 23261 ms · 2026-05-24T14:18:40.408647+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. New representations of the Hu-Meyer formulas and series expansion of iterated Stratonovich stochastic integrals with respect to components of a multidimensional Wiener process

    math.PR 2025-10 unverdicted novelty 6.0

    Derives new Hu-Meyer representations and verifies sufficient conditions for iterated Stratonovich integrals w.r.t. multidimensional Wiener process components using generalized multiple Fourier series.