pith. sign in

arxiv: 2502.09032 · v3 · pith:KPNQMNGFnew · submitted 2025-02-13 · 🧮 math.PR · math.CA

Fourier dimension of the graph of fractional Brownian motion with H ge 1/2

Chun-Kit Lai , Cheuk Yin Lee This is my paper

Pith reviewed 2026-05-23 03:49 UTC · model grok-4.3

classification 🧮 math.PR math.CA
keywords Fourier dimensionfractional Brownian motiongraphalmost surelyintegration by partsstable processesSalem set
0
0 comments X

The pith

The Fourier dimension of the graph of fractional Brownian motion with Hurst index greater than 1/2 is almost surely 1.

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

The paper proves that the graph of fractional Brownian motion with Hurst index H greater than 1/2 has Fourier dimension equal to 1 almost surely. This extends the corresponding result for standard Brownian motion and confirms part of a conjecture from 2014. The argument relies on a new combinatorial integration by parts formula that evaluates moments of the Fourier transform of the graph measure, together with Faà di Bruno's formula and the strong local nondeterminism of the process. The same approach also yields Fourier dimension 1 for the graphs of symmetric alpha-stable processes when alpha lies in [1,2], with the additional conclusion that the graph is a Salem set when alpha equals 1.

Core claim

We prove that the Fourier dimension of the graph of fractional Brownian motion with Hurst index greater than 1/2 is almost surely 1. This extends the result of Fraser and Sahlsten (2018) for the Brownian motion and confirms part of the conjecture of Fraser, Orponen and Sahlsten (2014). We introduce a combinatorial integration by parts formula to compute the moments of the Fourier transform of the graph measure. The proof of our main result is based on this integration by parts formula together with Faà di Bruno's formula and strong local nondeterminism of fractional Brownian motion. We also show that the graph of a symmetric alpha-stable process has Fourier dimension 1 almost surely when α ∈

What carries the argument

The combinatorial integration by parts formula that computes the moments of the Fourier transform of the graph measure.

Load-bearing premise

The combinatorial integration by parts formula correctly computes the moments of the Fourier transform of the graph measure.

What would settle it

A single sample path of fractional Brownian motion with Hurst index strictly greater than 1/2 whose graph has Fourier dimension strictly less than 1 would falsify the claim.

read the original abstract

We prove that the Fourier dimension of the graph of fractional Brownian motion with Hurst index greater than $1/2$ is almost surely 1. This extends the result of Fraser and Sahlsten (2018) for the Brownian motion and confirms part of the conjecture of Fraser, Orponen and Sahlsten (2014). We introduce a combinatorial integration by parts formula to compute the moments of the Fourier transform of the graph measure. The proof of our main result is based on this integration by parts formula together with Fa\`a di Bruno's formula and strong local nondeterminism of fractional Brownian motion. We also show that the graph of a symmetric $\alpha$-stable process has Fourier dimension 1 almost surely when $\alpha \in [1,2]$ and is a Salem set when $\alpha = 1$.

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 proves that the Fourier dimension of the graph of fractional Brownian motion with Hurst index H ≥ 1/2 is almost surely 1. This is achieved via a new combinatorial integration by parts formula for the moments of the Fourier transform of the graph measure, combined with Faà di Bruno's formula and the strong local nondeterminism of fBM. The result extends the H=1/2 case of Fraser-Sahlsten (2018) and addresses part of the Fraser-Orponen-Sahlsten (2014) conjecture. Analogous statements are proved for the graphs of symmetric α-stable processes (Fourier dimension 1 a.s. for α ∈ [1,2]; Salem for α=1).

Significance. If the central derivation holds, the result is significant: it resolves the Fourier dimension for the graphs of a wide class of fBMs, confirms a conjecture, and supplies a new combinatorial tool for moment estimates on random measures. The extension to stable processes adds independent value. The reliance on the established strong local nondeterminism property is a methodological strength.

major comments (2)
  1. [Introduction and the section presenting the combinatorial formula] The combinatorial integration by parts formula (introduced to compute moments of the Fourier transform of the graph measure): its derivation and range of applicability to the graph of fBM when H > 1/2 constitute the load-bearing step. The abstract states that the formula, together with Faà di Bruno and strong local nondeterminism, yields the result, but explicit verification that the formula produces the claimed moment decay (without hidden assumptions on the Hölder regularity or boundary terms) is required before the subsequent estimates can be accepted.
  2. [Main proof section] Proof of the main theorem (the integration-by-parts step in the moment estimates): the passage from the new formula to the required upper bounds on the Fourier moments must be checked for H > 1/2; any gap in justifying the vanishing of remainder terms or the applicability of the estimates under the graph's regularity would prevent the conclusion that the Fourier dimension equals 1 a.s.
minor comments (2)
  1. [Preliminaries] Clarify at the first use the precise definition of the graph measure and its Fourier transform, including the normalization constants.
  2. [Introduction] Add a short remark comparing the new combinatorial formula with existing integration-by-parts identities for Gaussian processes.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for highlighting the central role of the combinatorial integration-by-parts formula. We address the two major comments below.

read point-by-point responses

  1. Referee: [Introduction and the section presenting the combinatorial formula] The combinatorial integration by parts formula (introduced to compute moments of the Fourier transform of the graph measure): its derivation and range of applicability to the graph of fBM when H > 1/2 constitute the load-bearing step. The abstract states that the formula, together with Faà di Bruno and strong local nondeterminism, yields the result, but explicit verification that the formula produces the claimed moment decay (without hidden assumptions on the Hölder regularity or boundary terms) is required before the subsequent estimates can be accepted.

    Authors: The combinatorial integration by parts formula is derived in full in Section 3 via a direct combinatorial argument that applies to the graph measure for any Hurst index in (0,1). The derivation does not invoke Hölder regularity beyond the standard sample-path properties of fBM. In the proof of Theorem 4.1 we combine the formula with Faà di Bruno’s formula and verify the required moment decay explicitly; the boundary terms vanish because the test functions have compact support and the underlying measure is atomless. Strong local nondeterminism (which holds for all H ≥ 1/2) supplies the necessary variance bounds. We are prepared to insert a short clarifying paragraph in Section 3 that isolates the vanishing of remainders if the referee finds the current presentation insufficiently explicit. revision: partial

  2. Referee: [Main proof section] Proof of the main theorem (the integration-by-parts step in the moment estimates): the passage from the new formula to the required upper bounds on the Fourier moments must be checked for H > 1/2; any gap in justifying the vanishing of remainder terms or the applicability of the estimates under the graph's regularity would prevent the conclusion that the Fourier dimension equals 1 a.s.

    Authors: Section 4 carries out the passage from the combinatorial formula to the moment upper bounds. For H > 1/2 the graph is Hölder continuous of order H; this regularity is used only to control the size of increments when applying the nondeterminism estimates. Each remainder term arising after Faà di Bruno expansion is bounded by a direct Gaussian tail argument that is uniform in the range H ≥ 1/2. The same estimates that work for H = 1/2 extend immediately once the nondeterminism constant is adjusted; no additional vanishing arguments are required. We therefore see no gap that would invalidate the conclusion that the Fourier dimension is 1 almost surely. revision: no

Circularity Check

0 steps flagged

No circularity: new formula and external properties drive the proof

full rationale

The derivation introduces a combinatorial integration by parts formula (new to this paper) to handle moments of the graph measure Fourier transform, then applies Faà di Bruno's formula and the known strong local nondeterminism property of fBM (H > 1/2). These steps are independent of the target Fourier dimension result; the proof extends prior work by different authors (Fraser-Sahlsten 2018) without self-citation load-bearing or any reduction of the claimed dimension to a fitted quantity or self-defined input. No equations equate the result to its own construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the validity of the newly introduced combinatorial integration by parts formula and the domain property of strong local nondeterminism for fBM; no free parameters or invented entities are indicated in the abstract.

axioms (2)
  • domain assumption strong local nondeterminism of fractional Brownian motion
    Invoked explicitly in the abstract as a key ingredient for the proof of the main result.
  • standard math Faà di Bruno's formula applies to the composition in the Fourier transform moments
    Cited as part of the proof machinery in the abstract.

pith-pipeline@v0.9.0 · 5671 in / 1470 out tokens · 71869 ms · 2026-05-23T03:49:16.533347+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

50 extracted references · 50 canonical work pages

  1. [1]

    Adler, Hausdorff dimension and G aussian fields , Ann

    Robert J. Adler, Hausdorff dimension and G aussian fields , Ann. Probability 5 (1977), no. 1, 145--151. 426123

    work page 1977

  2. [2]

    Jean Bourgain and Semyon Dyatlov, Fourier dimension and spectral gaps for hyperbolic surfaces, Geom. Funct. Anal. 27 (2017), no. 4, 744--771. 3678500

    work page 2017

  3. [3]

    Berman, Local nondeterminism and local times of G aussian processes , Indiana Univ

    Simeon M. Berman, Local nondeterminism and local times of G aussian processes , Indiana Univ. Math. J. 23 (1973/74), 69--94. 317397

    work page 1973

  4. [4]

    R. M. Blumenthal and R. K. Getoor, The dimension of the set of zeros and the graph of a symmetric stable process, Illinois J. Math. 6 (1962), 308--316. 138134

    work page 1962

  5. [5]

    Francesca Biagini, Yaozhong Hu, Bernt ksendal, and Tusheng Zhang, Stochastic calculus for fractional B rownian motion and applications , Probability and its Applications (New York), Springer-Verlag London, Ltd., London, 2008. 2387368

    work page 2008

  6. [6]

    DuPreez, Joint continuity of G aussian local times , Ann

    Jack Cuzick and Johannes P. DuPreez, Joint continuity of G aussian local times , Ann. Probab. 10 (1982), no. 3, 810--817. 659550

    work page 1982

  7. [7]

    Xinxin Chen, Yong Han, Yanqi Qiu, and Zipeng Wang, Harmonic analysis of M andelbrot cascades--in the context of vector-valued martingales , https://arxiv.org/pdf/2409.13164

    work page arXiv

  8. [8]

    Changhao Chen, Bing Li, and Ville Suomala, Fourier dimension of M andelbrot multiplicative cascade , https://arxiv.org/pdf/2409.13455

    work page arXiv

  9. [9]

    Conway, A course in functional analysis, second ed., Graduate Texts in Mathematics, vol

    John B. Conway, A course in functional analysis, second ed., Graduate Texts in Mathematics, vol. 96, Springer-Verlag, New York, 1990. 1070713

    work page 1990

  10. [10]

    Davenport, P

    H. Davenport, P. Erd o s, and W. J. LeVeque, On W eyl's criterion for uniform distribution , Michigan Math. J. 10 (1963), 311--314. 153656

    work page 1963

  11. [11]

    Dexter Dysthe and Chun-Kit Lai, Hausdorff and F ourier dimension of graph of continuous additive processes , Stochastic Process. Appl. 155 (2023), 355--392. 4509492

    work page 2023

  12. [12]

    Fredrik Ekstr\"om, Fourier dimension of random images, Ark. Mat. 54 (2016), no. 2, 455--471. 3546361

    work page 2016

  13. [13]

    Kenneth Falconer, Fractal geometry, John Wiley & Sons, Ltd., Chichester, 1990, Mathematical foundations and applications. 1102677

    work page 1990

  14. [14]

    1855, 479--80

    Francesco Fa \`a di Bruno, Francesco, Sullo sviluppo delle funzioni, Annali di scienze matematiche e fisiche 6 (1855), no. 1855, 479--80

    work page

  15. [15]

    Pure Appl

    Francesco Fa \`a di Bruno, Note sur une nouvelle formule de calcul diff \'e rentiel , Quarterly J. Pure Appl. Math 1 (1857), no. 359-360, 12

    work page

  16. [16]

    Robert Fraser and Kyle Hambrook, Explicit S alem sets in R ^n , Adv. Math. 416 (2023), Paper No. 108901, 23. 4548424

    work page 2023

  17. [17]

    Fraser, Tuomas Orponen, and Tuomas Sahlsten, On F ourier analytic properties of graphs , Int

    Jonathan M. Fraser, Tuomas Orponen, and Tuomas Sahlsten, On F ourier analytic properties of graphs , Int. Math. Res. Not. IMRN (2014), no. 10, 2730--2745. 3214283

    work page 2014

  18. [18]

    Fraser, The F ourier spectrum and sumset type problems , Math

    Jonathan M. Fraser, The F ourier spectrum and sumset type problems , Math. Ann. 390 (2024), no. 3, 3891--3930. 4803466

    work page 2024

  19. [19]

    Fraser and Tuomas Sahlsten, On the F ourier analytic structure of the B rownian graph , Anal

    Jonathan M. Fraser and Tuomas Sahlsten, On the F ourier analytic structure of the B rownian graph , Anal. PDE 11 (2018), no. 1, 115--132. 3707292

    work page 2018

  20. [20]

    Michael Hardy, Combinatorics of partial derivatives, Electron. J. Combin. 13 (2006), no. 1, Research Paper 1, 13. 2200529

    work page 2006

  21. [21]

    Thomas Jordan and Tuomas Sahlsten, Fourier transforms of G ibbs measures for the G auss map , Math. Ann. 364 (2016), no. 3-4, 983--1023. 3466857

    work page 2016

  22. [22]

    Jean-Pierre Kahane, Sur les mauvaises r\' e partitions modulo 1 , Ann. Inst. Fourier (Grenoble) 14 (1964), no. fasc. 2, 519--526. 174545

    work page 1964

  23. [23]

    5, Cambridge University Press, Cambridge, 1985

    , Some random series of functions, second ed., Cambridge Studies in Advanced Mathematics, vol. 5, Cambridge University Press, Cambridge, 1985. 833073

    work page 1985

  24. [24]

    Kahane, Fractals and random measures, Bull

    J.-P. Kahane, Fractals and random measures, Bull. Sci. Math. 117 (1993), no. 1, 153--159. 1205416

    work page 1993

  25. [25]

    Kaufman, Continued fractions and F ourier transforms , Mathematika 27 (1980), no

    R. Kaufman, Continued fractions and F ourier transforms , Mathematika 27 (1980), no. 2, 262--267. 610711

    work page 1980

  26. [26]

    39 (1981), no

    , On the theorem of J arn\'ik and B esicovitch , Acta Arith. 39 (1981), no. 3, 265--267. 640914

    work page 1981

  27. [27]

    Davar Khoshnevisan and Cheuk Yin Lee, On the passage times of self-similar G aussian processes on curved boundaries , arXiv preprint arXiv:2506.15949 (2025)

    work page arXiv 2025

  28. [28]

    2, 1504--1534

    Kamran Kalbasi and Thomas Mountford, On the probability distribution of the local times of diagonally operator-self-similar G aussian fields with stationary increments , Bernoulli 26 (2020), no. 2, 1504--1534. 4058376

    work page 2020

  29. [29]

    Davar Khoshnevisan, Dongsheng Wu, and Yimin Xiao, Sectorial local non-determinism and the geometry of the B rownian sheet , Electron. J. Probab. 11 (2006), no. 32, 817--843. 2261054

    work page 2006

  30. [30]

    1, 277--306

    Cheuk Yin Lee, The H ausdorff measure of the range and level sets of G aussian random fields with sectorial local nondeterminism , Bernoulli 28 (2022), no. 1, 277--306. 4337706

    work page 2022

  31. [31]

    Izabella aba and Malabika Pramanik, Arithmetic progressions in sets of fractional dimension, Geom. Funct. Anal. 19 (2009), no. 2, 429--456. 2545245

    work page 2009

  32. [32]

    Yiyu Liang and Malabika Pramanik, Fourier dimension and avoidance of linear patterns, Adv. Math. 399 (2022), Paper No. 108252, 50. 4384610

    work page 2022

  33. [33]

    1, 523--550

    Cheuk Yin Lee and Yimin Xiao, Chung-type law of the iterated logarithm and exact moduli of continuity for a class of anisotropic G aussian random fields , Bernoulli 29 (2023), no. 1, 523--550. 4497257

    work page 2023

  34. [34]

    , Local times of anisotropic G aussian random fields and stochastic heat equation , arXiv preprint arXiv:2308.13732 (2023)

    work page arXiv 2023

  35. [35]

    150, Cambridge University Press, Cambridge, 2015

    Pertti Mattila, Fourier analysis and H ausdorff dimension , Cambridge Studies in Advanced Mathematics, vol. 150, Cambridge University Press, Cambridge, 2015. 3617376

    work page 2015

  36. [36]

    Mishura, Stochastic calculus for fractional B rownian motion and related processes , Lecture Notes in Mathematics, vol

    Yuliya S. Mishura, Stochastic calculus for fractional B rownian motion and related processes , Lecture Notes in Mathematics, vol. 1929, Springer-Verlag, Berlin, 2008. 2378138

    work page 1929

  37. [37]

    Pitt, Local nondeterminism and H ausdorff dimension , Seminar on stochastic processes, 1986 ( C harlottesville, V a., 1986), Progr

    Ditlev Monrad and Loren D. Pitt, Local nondeterminism and H ausdorff dimension , Seminar on stochastic processes, 1986 ( C harlottesville, V a., 1986), Progr. Probab. Statist., vol. 13, Birkh\"auser Boston, Boston, MA, 1987, pp. 163--189. 902433

    work page 1986

  38. [38]

    Mosquera and Pablo S

    Carolina A. Mosquera and Pablo S. Shmerkin, Self-similar measures: asymptotic bounds for the dimension and F ourier decay of smooth images , Ann. Acad. Sci. Fenn. Math. 43 (2018), no. 2, 823--834. 3839838

    work page 2018

  39. [39]

    Fourier Anal

    Safari Mukeru, The zero set of fractional B rownian motion is a S alem set , J. Fourier Anal. Appl. 24 (2018), no. 4, 957--999. 3843846

    work page 2018

  40. [40]

    David Nualart, The M alliavin calculus and related topics , second ed., Probability and its Applications (New York), Springer-Verlag, Berlin, 2006. 2200233

    work page 2006

  41. [41]

    Wahrscheinlichkeitstheorie und Verw

    Steven Orey, Gaussian sample functions and the H ausdorff dimension of level crossings , Z. Wahrscheinlichkeitstheorie und Verw. Gebiete 15 (1970), 249--256. 279882

    work page 1970

  42. [42]

    Pitt, Local times for G aussian vector fields , Indiana Univ

    Loren D. Pitt, Local times for G aussian vector fields , Indiana Univ. Math. J. 27 (1978), no. 2, 309--330. 471055

    work page 1978

  43. [43]

    Pablo Shmerkin, Salem sets with no arithmetic progressions, Int. Math. Res. Not. IMRN (2017), no. 7, 1929--1941. 3658188

    work page 2017

  44. [44]

    Pablo Shmerkin and Ville Suomala, Spatially independent martingales, intersections, and applications, Mem. Amer. Math. Soc. 251 (2018), no. 1195, v+102. 3756896

    work page 2018

  45. [45]

    176 (2006), no

    Narn-Rueih Shieh and Yimin Xiao, Images of G aussian random fields: S alem sets and interior points , Studia Math. 176 (2006), no. 1, 37--60. 2263961

    work page 2006

  46. [46]

    S. J. Taylor, The H ausdorff -dimensional measure of B rownian paths in n -space , Proc. Cambridge Philos. Soc. 49 (1953), 31--39. 52719

    work page 1953

  47. [47]

    Fourier Anal

    Dongsheng Wu and Yimin Xiao, Geometric properties of fractional B rownian sheets , J. Fourier Anal. Appl. 13 (2007), no. 1, 1--37. 2296726

    work page 2007

  48. [48]

    Yimin Xiao, Strong local nondeterminism and sample path properties of G aussian random fields , Asymptotic theory in probability and statistics with applications, Adv. Lect. Math. (ALM), vol. 2, Int. Press, Somerville, MA, 2008, pp. 136--176. 2466984

    work page 2008

  49. [49]

    1962, Springer, Berlin, 2009, pp

    , Sample path properties of anisotropic G aussian random fields , A minicourse on stochastic partial differential equations, Lecture Notes in Math., vol. 1962, Springer, Berlin, 2009, pp. 145--212. 2508776

    work page 1962

  50. [50]

    255--288

    , Recent developments on fractal properties of G aussian random fields , Further developments in fractals and related fields, Trends Math., Birkh\"auser/Springer, New York, 2013, pp. 255--288. 3184196

    work page 2013