Pith. sign in

REVIEW 5 minor 54 references

Sharp moduli of continuity for Gaussian fields and stochastic PDEs via correlation bounds

T0 review · 0 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read This paper proves exact moduli of continuity for Gaussian fields from correlation bounds alone, and settles the spatial modulus of a critical linear SPDE at $2H\alpha=3$.

desk verdict Strong new framework proves the critical SPDE spatial modulus that SLND couldn't reach; the only real caveat is temporal results leaning on an unpublished preprint. read the letter →

arxiv 2607.20956 v1 pith:7U47DTP6 submitted 2026-07-23 math.PR

classification math.PR MSC 60G1560G6060G1760H15
keywords modulusofcontinuitylawtheiteratedlogarithmGaussianrandomfieldsstochasticPDEscorrelationboundslocalizationestimatesdecorrelationlocalnondeterminism
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper develops a general method for exact moduli of continuity of Gaussian random fields, requiring only variance two-sided bounds and a correlation bound on pairwise increments instead of strong local nondeterminism, stationarity, or spectral representations. It claims that, under these bounds, every anisotropic Gaussian field has an exact uniform modulus of continuity of the form $\phi(t-s)\sqrt{\log[1/\psi(\phi(t-s))]}$ with a finite positive constant, and an analogous local modulus with $\ell(1/\psi(\phi(t-s)))$ for $\ell=\log\log$ or $\ell=\log$. As an application, it solves an open problem for linear stochastic PDEs in the critical case $2H\alpha=3$, where the solution is $C^{1-}$ in space and strong local nondeterminism is unavailable, proving that the spatial uniform modulus is $|x-y|\log(1/|x-y|)$ up to a constant and that the local spatial modulus carries an extra $\sqrt{\log\log\log(1/|x-x_0|)}$ factor. The proofs rest on new localization estimates for the SPDE increments that force decorrelation at small power-of-$h$ separations.

What carries the argument

The central object is the scale function $\psi(r)=\prod_{i=1}^N\phi_i^{-1}(r)$, which measures the 'number of increments' at scale $r$ through metric entropy; the key hypothesis is a correlation bound on pairwise increments, equivalently $\mathrm{Var}(X(t)-X(s)\mid X(t')-X(s'))\ge C_0\,\mathrm{Var}(X(t)-X(s))$ whenever the two increments are separated by at least $\psi^{-1}(A_0(\psi(h))^\rho)$ (Assumption 2.3, Lemma 4.4). This correlation bound lets Slepian's lemma compare the normalized increments with a one-factor Gaussian model $Z_i=(1-C_0)^{1/4}\xi_0+(1-\sqrt{1-C_0})^{1/2}\xi_i$, producing the lower bound on the modulus constant from the Gaussian tail of $\max_i\xi_i$; the upper bound comes from a Dudley-type entropy integral over $\psi$. For the SPDE, the same correlation condition is verified by localizing each spatial increment to the parabolic window $[t_0-h^{\rho\alpha},t_0]\times[x-h^\rho,x+h^\rho]$; Proposition 7.4 shows the localization error is at most $K\rho\,h^2\log(1/h)$ using the fractional heat-kernel gradient bound $|\partial_yG(s,y)|\lesssim s|y|^{-2-\alpha}$, and Proposition 7.5 converts this into decorrelation of increments separated by $2r^\rho$.

What would settle it

For the SPDE, compute directly the covariance of $\Delta_h u(t_0,x)$ and $\Delta_h u(t_0,x')$ at separation $|x-x'|=2r^\rho$ using the spectral formula (7.18); if for some $\rho<1$ this covariance exceeds $\sqrt{1-C_0}\,\sigma(h)^2$ for a fixed $C_0>0$ as $h\to0$, then the decorrelation Proposition 7.5 fails and the positivity of $K_3$ would be in doubt. For the general theorem, construct a Gaussian field satisfying Assumptions 2.1 and 2.2 but with increments at separation $\psi^{-1}(A_0(\psi(h))^\rho)$ having covariance bounded only by $\sqrt{1-C_0+\delta}$ for a fixed $\delta>0$; then the lower-bound argument in Theorem 2.4 gives $C\ge C_1[2\rho(1-\sqrt{\delta})]^{1/2}$, and for $\delta$ close to $1$ this can fail to be positive even though the upper bound holds.

Watch

Extended reading notes

Core claim

For a centered Gaussian field $X$ on a compact rectangle, suppose $\|X(t)-X(s)\|_2\asymp\phi(t-s)$ with $\phi_i$ regularly varying of index $\theta_i\in[0,1]$, and suppose increments separated by $\psi^{-1}(A_0(\psi(h))^\rho)$ are effectively independent in the sense that their covariance is at most $\sqrt{1-C_0}$ times the product of their $\mathrm{L}^2$ norms (Assumption 2.3, equivalent to a conditional-variance lower bound). Then Theorem 2.4 gives $\lim_{h\to0+}\sup_{t,s:0<\phi(t-s)\le h}|X(t)-X(s)|/(\phi(t-s)\sqrt{\log[1/\psi(\phi(t-s))]})=C$ almost surely for a finite positive constant $C$, with matching upper and lower bounds in terms of $C_0,\rho$ and the metric-entropy constants. For the SPDE $\partial_t u=-(-\Delta)^{\alpha/2}u+\dot W$ with $\dot W$ fractional in time of Hurst index $H$ and white in space, in the critical regime $\alpha\in(3/2,2], 2H\alpha=3$ on $\mathbb{R}$, Theorem 1.5 proves $\lim_{h\to0+}\sup_{x,y\in J:0<|x-y|\le h}|u(t_0,x)-u(t_0,y)|/(|x-y|\log(1/|x-y|))=K_3$ a.s., and Theorem 1.6 proves the local modulus with denominator $|x-x_0|\sqrt{\log(1/|x-x_0|)\log\log\log(1/|x-x_0|)}=K_6$.

Load-bearing premise

Everything rests on the correlation bound that increments at distance $h$ become nearly independent as soon as their starting points are separated by a small power of $h$; if that bound fails or the localization error in the SPDE is not proportional to $\rho$, the lower bound on the modulus constant collapses.

Editorial extensions

If this is right

  • The uniform spatial modulus of the critical SPDE solution at a fixed time is exactly $K_3|x-y|\log(1/|x-y|)$ for some finite positive $K_3$, so the solution's spatial sample paths are as rough as the borderline between continuous and differentiable at every scale.
  • At a fixed space-time point, spatial increments satisfy a local law of the iterated logarithm with denominator $|x-x_0|\sqrt{\log(1/|x-x_0|)\log\log\log(1/|x-x_0|)}$, a rate not captured by earlier methods.
  • Temporal and joint moduli follow: temporal increments at fixed $x_0$ have modulus $|t-s|^{1/\alpha}\sqrt{\log(1/|t-s|)}$ uniformly and $\sqrt{\log\log}$ locally, and the joint field has modulus $\phi(z,z')\sqrt{\log(1/\phi(z,z'))}$.
  • Any Gaussian field whose increments satisfy the correlation bound has an exact modulus constant that is finite and positive; the framework therefore gives the same conclusions as SLND-based results whenever SLND holds.
  • Gaussian Volterra processes with slowly varying variance function $\phi(r)=L(\log(1/r))/(\log(1/r))^a$, $a>1/2$, have the same uniform and local modulus functions, with a common constant.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same correlation-bound route could identify exact phase transitions for other SPDEs where SLND is open, such as stochastic wave equations or fractional-noise equations in higher dimensions, by verifying that the localization error is proportional to $\rho$.
  • The need for the doubly-exponential scale $h_n=\exp(-e^{n^\gamma})$ in the local spatial modulus suggests that fixed-point spatial increments of the critical SPDE decorrelate only after enormous rescaling; this may reflect an 'approximate derivative' whose variance grows like $\sqrt{n\log\log n}$, connecting to the harmonizable derivative $D(n)$.
  • For non-critical cases $\theta_2\in(0,1)$ with SLND, the same framework should recover the same constants, suggesting the correlation-bound condition is the 'right' hypothesis and SLND is only a sufficient tool.
  • Testing the covariance numerically at separation $h^\rho$ for small $\rho$ would provide a direct verification of Proposition 7.5, and could suggest the optimal $\rho(\varepsilon)$ tradeoff.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

0 major / 5 minor

Summary. The paper proposes a general framework for exact uniform and local moduli of continuity of Gaussian random fields, replacing the usual strong local nondeterminism (SLND) assumption by correlation bounds on pairwise increments (Assumptions 2.3 and 2.6). The main general results are Theorems 2.4 and 2.7 for anisotropic fields, with scalar corollaries Theorems 1.1 and 1.2. The central application is to the linear SPDE (1.9) in the critical case 2Hα=3, where the solution is spatially C^{1-} and SLND is not available. The authors prove sharp spatial, temporal, and joint moduli of continuity (Theorem 1.5) and sharp local moduli (Theorem 1.6). The key new ingredient is a set of localization estimates for spatial increments of the solution (Proposition 7.4), which yield decorrelation of sufficiently separated increments (Proposition 7.5). A short final section applies the framework to Gaussian Volterra processes with slowly varying variance functions.

Significance. If the main results are correct, this is a substantial contribution. The framework provides a route to sharp moduli of continuity that does not require SLND, stationarity, or spectral representations, and it solves an open problem for the critical SPDE case 2Hα=3. The localization and decorrelation estimates in Section 7 are likely to be of independent interest, and the sharp limit statements (1.10) and (1.13) are crisp falsifiable predictions. The proofs are detailed and self-contained for the central spatial claims; constants are not fitted but are shown to be positive almost-sure limits. The dependence of the temporal parts on the preprint [7] is a caveat, but it does not affect the main spatial critical-case result.

minor comments (5)
  1. [§7.2, proof of (1.12)] The displayed lower bound for K5 says the temporal restriction gives a limit equal to αK4, but the denominator reads |t-s|^{1/α}√log(1/|t-s|^{1/α}) = α^{-1/2}|t-s|^{1/α}√log(1/|t-s|), so the limit is α^{1/2}K4, not αK4. Positivity is unaffected, but the displayed equality should be corrected.
  2. [§8, Lemma 8.5(iii)] The step Var(X(t_n)-X(t_0) | X(t_m)-X(t_0)) ≥ φ²(t_n-t_m) is justified by referring to one-sided SLND, but Lemma 8.2(ii) applies only when the conditioning points precede the evaluation point t_n; here t_m > t_n. The statement can be repaired by applying Lemma 8.2(ii) to X(t_m) conditional on X(t_n), X(t_0) and then using the Gaussian identity Var(Y | Y+W) = Var(Y) Var(W | Y)/Var(Y+W) with Y = X(t_n)-X(t_0) and W = X(t_m)-X(t_n); this gives the claimed lower bound up to an absolute constant. As written, the proof is incomplete.
  3. [§7.2 and §7.3] The proofs of (1.11), (1.14), and hence the lower bounds for K5 and K8, invoke Theorem 1.3 of the unpublished e-print [7] for temporal SLND. This does not affect the spatial results (1.10) or (1.13), but the authors should either provide a proof of the needed temporal SLND statement or confirm that [7] is publicly available and acceptable for citation.
  4. [§7.1, Lemma 7.2 proof] In the scaling argument after the Fourier representation, the text says only that ξ is changed to δ^{-1/α}ξ, but the displayed integral also requires the simultaneous scaling τ ↦ δ^{-1}τ in order for the denominator to become |τ|²+|ξ|^{2α} and for the exponential factor to become exp(-(iτ+|ξ|^α)(t-δ)/δ). The final estimates are correct, but the presentation should state both scalings.
  5. [§1 and §2, notation] The phrase 'C^{1-} in space' in the abstract and introduction is informal; since the paper elsewhere uses θ_2^- notation for Hölder exponents, it would be clearer to say that the spatial sample paths are Hölder continuous of every order below 1.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reduction: the SPDE decorrelation argument is independently verified and the a.s. constants are not fitted; only auxiliary temporal SLND is imported from a co-authored preprint.

full rationale

The paper's central claim (Theorem 1.5(1.10)) is not circular. Condition (1.4) is a hypothesis of Theorem 1.1, and for the SPDE it is verified in Section 7.2 from Proposition 7.4: with delta = h^{rho alpha} and epsilon = h^rho, Lemmas 7.2 and 7.3 give ||Delta^h_B u||_2^2 <= K rho h^2 log(1/h), and the covariance decomposition plus Cauchy-Schwarz yields a factor sqrt(rho) that is made <= sqrt(1-C0) C1^2 by choosing rho small. This is a genuine derivation, not a restatement of the desired modulus. The constants K3 and K6 are a.s. limits proved finite and positive; they are not fitted to data. Lemma 4.4 merely translates the conditional-variance condition into the correlation bound, and Proposition 3.1 shows LND implies the condition, but the framework does not require the converse. The only author-overlapping citation is [7], used for temporal SLND in (1.11)/(1.14) and hence for positivity of K5/K8; this is peripheral to the main spatial critical-case result (1.10) and to the local spatial result (1.13). No equation is shown to equal another by construction, and no fitted parameter is renamed as a prediction.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

The central proof rests on standard Gaussian process tools and on the new correlation conditions, plus several results imported from prior work: the variance equivalence for the SPDE, the temporal SLND citation, and the heat kernel gradient estimate. No free parameters are fitted to data; the constants in the moduli are a.s. limits shown to be finite and positive.

assumptions (8)
  • domain assumption Assumption 2.1: the Gaussian field's increment variance is comparable to a regular-varying function ϕ at small scales.
    Core hypothesis of the general theorems; for the SPDE this is Lemma 7.1 from prior literature.
  • domain assumption Assumption 2.2 and 2.5: metric entropy integral bounds (2.4)-(2.5) and (2.7)-(2.8).
    Ensures finite expected suprema and yields the correct normalizing rate; verified using regular variation.
  • domain assumption Assumption 2.3 and 2.6, i.e., pairwise increment correlation bounds (1.4) and (1.7).
    The new replacement for SLND; proven for the SPDE's spatial process via localization, and for time marginals via [7, Theorem 1.3].
  • standard math Dudley's metric entropy bound, Slepian's lemma, Borell's inequality, and the Gaussian zero-one law.
    Standard tools for upper and lower bounds on suprema of Gaussian processes; cited from [1,6,12,35,48].
  • standard math Regular variation theory, especially the asymptotic inversion properties (4.2)-(4.3).
    Used throughout to relate φ, φ^{-1}, ψ and to verify integral conditions.
  • domain assumption The SPDE (7.1) has a pointwise mild solution under (7.5), with increment variance equivalence given by Lemma 7.1.
    Baseline SPDE regularity, taken from [7,16,29]; needed to match Assumption 2.1.
  • domain assumption Temporal SLND for {u(t,x0)}_{t∈I} from [7, Theorem 1.3].
    Used for the temporal moduli (1.11) and (1.14); this is a supporting citation, not part of the new spatial decorrelation proof.
  • domain assumption Heat kernel gradient bound |∂_y G(s,y)| ≲ s |y|^{-2-α} (7.13).
    Used in Lemma 7.3 to estimate the far spatial contribution to increments; cited from [8, Lemma 2.2].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sharp moduli of continuity for Gaussian fields and stochastic PDEs via correlation bounds." pith.science (2026). https://pith.science/paper/7U47DTP6

@misc{pith2026260720956,
  author       = {Pith},
  title        = {Pith review of: Sharp moduli of continuity for Gaussian fields and stochastic PDEs via correlation bounds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7U47DTP6}},
  note         = {Machine review of arXiv:2607.20956}
}
abstract

Exact uniform and local moduli of continuity for anisotropic Gaussian random fields are established under a general framework based on correlation bounds for pairwise increments. This framework does not necessarily require strong local nondeterminism (SLND), stationarity of increments, or spectral-type representations. As an application, we solve an open problem about sharp moduli of continuity for a class of linear stochastic PDEs in a particular case where the solution is $C^{1-}$ in space and SLND is not available. In this case, we establish decorrelation of the spatial increments via new localization estimates, which may be of independent interest. We also briefly discuss an example about Gaussian Volterra processes.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 47 canonical work pages

  1. [7]

    L. Chen, C. Y. Lee, and P. Xia. Strong local nondeterminism for stochastic time-fractional slow and fast diffusion equations.E-print available at arXiv:2602.05317., 2026

  2. [1]

    R. J. Adler and J. E. Taylor.Random fields and geometry. Springer Monographs in Mathe- matics. Springer, New York, 2007

  3. [2]

    Allouba and Y

    H. Allouba and Y. Xiao. L-Kuramoto-Sivashinsky SPDEs vs. time-fractional SPIDEs: exact continuity and gradient moduli, 1/2-derivative criticality, and laws.J. Differential Equations, 263(2):1552–1610, 2017

  4. [3]

    R. M. Balan. Linear SPDEs driven by stationary random distributions.J. Fourier Anal. Appl., 18(6):1113–1145, 2012

  5. [4]

    S. M. Berman. Local nondeterminism and local times of Gaussian processes.Indiana Univ. Math. J., 23:69–94, 1973/74

  6. [5]

    N. H. Bingham, C. M. Goldie, and J. L. Teugels.Regular variation, volume 27 ofEncyclopedia of Mathematics and its Applications. Cambridge University Press, Cambridge, 1987

  7. [6]

    C. Borell. The Brunn-Minkowski inequality in Gauss space.Invent. Math., 30(2):207–216, 1975

  8. [8]

    Chen and X

    Z.-Q. Chen and X. Zhang. Heat kernels and analyticity of non-symmetric jump diffusion semigroups.Probab. Theory Related Fields, 165(1-2):267–312, 2016

Show all 54 references
  1. [9]

    Cuzick and J

    J. Cuzick and J. P. DuPreez. Joint continuity of Gaussian local times.Ann. Probab., 10(3):810–817, 1982. 32 C.Y. LEE AND T. ZHANG

  2. [10]

    R. C. Dalang. Extending the martingale measure stochastic integral with applications to spatially homogeneous s.p.d.e.’s.Electron. J. Probab., 4:no. 6, 29, 1999

  3. [11]

    R. C. Dalang, C. Mueller, and Y. Xiao. Polarity of points for Gaussian random fields.Ann. Probab., 45(6B):4700–4751, 2017

  4. [12]

    R. M. Dudley. The sizes of compact subsets of Hilbert space and continuity of Gaussian processes.J. Functional Analysis, 1:290–330, 1967

  5. [13]

    Fernique

    X. Fernique. Continuit´ e des processus Gaussiens.C. R. Acad. Sci. Paris, 258:6058–6060, 1964

  6. [14]

    Foondun, D

    M. Foondun, D. Khoshnevisan, and P. Mahboubi. Analysis of the gradient of the solution to a stochastic heat equation via fractional Brownian motion.Stoch. Partial Differ. Equ. Anal. Comput., 3(2):133–158, 2015

  7. [15]

    Y. Guo, J. Song, R. Wang, and Y. Xiao. Sample path properties and small ball probabilities for stochastic fractional diffusion equations.J. Differential Equations, 446:Paper No. 113604, 56, 2025

  8. [16]

    Herrell, R

    R. Herrell, R. Song, D. Wu, and Y. Xiao. Sharp space-time regularity of the solution to stochastic heat equation driven by fractional-colored noise.Stoch. Anal. Appl., 38(4):747– 768, 2020

  9. [17]

    Hinojosa-Calleja

    A. Hinojosa-Calleja. Exact uniform modulus of continuity forq-isotropic Gaussian random fields.Statist. Probab. Lett., 197:Paper No. 109813, 9, 2023

  10. [18]

    Hu and C

    J. Hu and C. Y. Lee. On the spatio-temporal increments of nonlinear parabolic SPDEs and the open KPZ equation.To appear in Ann. Appl. Probab. E-print available at arXiv:2508.05032., 2025

  11. [19]

    Khoshnevisan and C

    D. Khoshnevisan and C. Y. Lee. On the passage times of self-similar Gaussian processes on curved boundaries.Bernoulli, 32(3):2098–2122, 2026

  12. [20]

    Khoshnevisan, C

    D. Khoshnevisan, C. Y. Lee, F. Pu, and Y. Xiao. Uniform dimension theorems for parabolic SPDEs.E-print available at arXiv:2511.04938., 2025

  13. [21]

    N. Kˆ ono. On the modulus of continuity of sample functions of Gaussian processes.J. Math. Kyoto Univ., 10:493–536, 1970

  14. [22]

    Lai and C

    C.-K. Lai and C. Y. Lee. Fourier dimension of the graph of fractional Brownian motion with H≥1/2.E-print available at arXiv:2502.09032., 2025

  15. [23]

    X. Lan, D. Marinucci, and Y. Xiao. Strong local nondeterminism and exact modulus of continuity for spherical Gaussian fields.Stochastic Process. Appl., 128(4):1294–1315, 2018

  16. [24]

    Lan and Y

    X. Lan and Y. Xiao. Regularity properties of the solution to a stochastic heat equation driven by a fractional Gaussian noise onS 2.J. Math. Anal. Appl., 476(1):27–52, 2019

  17. [25]

    C. Y. Lee. The Hausdorff measure of the range and level sets of Gaussian random fields with sectorial local nondeterminism.Bernoulli, 28(1):277–306, 2022

  18. [26]

    C. Y. Lee. Local nondeterminism and local times of the stochastic wave equation driven by fractional-colored noise.J. Fourier Anal. Appl., 28(2):Paper No. 26, 38, 2022

  19. [27]

    C. Y. Lee and S. Tindel. On the Fourier dimension of fractional Brownian graphs.E-print available at arXiv:2510.09818., 2025

  20. [28]

    C. Y. Lee and Y. Xiao. Local nondeterminism and the exact modulus of continuity for sto- chastic wave equation.Electron. Commun. Probab., 24:Paper No. 52, 8, 2019

  21. [29]

    C. Y. Lee and Y. Xiao. Chung-type law of the iterated logarithm and exact moduli of conti- nuity for a class of anisotropic Gaussian random fields.Bernoulli, 29(1):523–550, 2023

  22. [30]

    C. Y. Lee and Y. Xiao. Local times of anisotropic Gaussian random fields and stochastic heat equation.To appear in Ann. Inst. H. Poincar´ e Probab. Statist. E-print available at arXiv:2308.13732., 2023

  23. [31]

    Y. Li, W. Wang, and Y. Xiao. Exact moduli of continuity for operator-scaling Gaussian random fields.Bernoulli, 21(2):930–956, 2015

  24. [32]

    M. B. Marcus. H¨ older conditions for Gaussian processes with stationary increments.Trans. Amer. Math. Soc., 134:29–52, 1968

  25. [33]

    M. B. Marcus. H¨ older conditions for continuous Gaussian processes.Osaka Math. J., 7:483– 493, 1970

  26. [34]

    M. B. Marcus and J. Rosen. Moduli of continuity of local times of strongly symmetric Markov processes via Gaussian processes.J. Theoret. Probab., 5(4):791–825, 1992

  27. [35]

    M. B. Marcus and J. Rosen.Markov processes, Gaussian processes, and local times, volume 100 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 2006. SHARP MODULI OF CONTINUITY 33

  28. [36]

    M. B. Marcus and L. A. Shepp. Continuity of Gaussian processes.Trans. Amer. Math. Soc., 151:377–391, 1970

  29. [37]

    M. B. Marcus and L. A. Shepp. Sample behavior of Gaussian processes. InProceedings of the Sixth Berkeley Symposium on Mathematical Statistics and Probability (Univ. California, Berkeley, Calif., 1970/1971), Vol. II: Probability theory, pages 423–441. Univ. California Press, Be...

  30. [38]

    Mattila.Geometry of sets and measures in Euclidean spaces, volume 44 ofCambridge Studies in Advanced Mathematics

    P. Mattila.Geometry of sets and measures in Euclidean spaces, volume 44 ofCambridge Studies in Advanced Mathematics. Cambridge University Press, Cambridge, 1995. Fractals and rectifiability

  31. [39]

    M. M. Meerschaert, W. Wang, and Y. Xiao. Fernique-type inequalities and moduli of con- tinuity for anisotropic Gaussian random fields.Trans. Amer. Math. Soc., 365(2):1081–1107, 2013

  32. [40]

    Mocioalca and F

    O. Mocioalca and F. Viens. Skorohod integration and stochastic calculus beyond the fractional Brownian scale.J. Funct. Anal., 222(2):385–434, 2005

  33. [41]

    Monrad and L

    D. Monrad and L. D. Pitt. Local nondeterminism and Hausdorff dimension. InSeminar on stochastic processes, 1986 (Charlottesville, Va., 1986), volume 13 ofProgr. Probab. Statist., pages 163–189. Birkh¨ auser Boston, Boston, MA, 1987

  34. [42]

    M. Nisio. On the extreme values of Gaussian processes.Osaka Math. J., 4:313–326, 1967

  35. [43]

    Orey and W

    S. Orey and W. E. Pruitt. Sample functions of theN-parameter Wiener process.Ann. Prob- ability, 1(1):138–163, 1973

  36. [44]

    L. D. Pitt. Local times for Gaussian vector fields.Indiana Univ. Math. J., 27(2):309–330, 1978

  37. [45]

    Revuz and M

    D. Revuz and M. Yor.Continuous martingales and Brownian motion, volume 293 of Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathemati- cal Sciences]. Springer-Verlag, Berlin, second edition, 1994

  38. [46]

    Seneta.Regularly varying functions, volume Vol

    E. Seneta.Regularly varying functions, volume Vol. 508 ofLecture Notes in Mathematics. Springer-Verlag, Berlin-New York, 1976

  39. [47]

    Sirao and H

    T. Sirao and H. Watanabe. On the upper and lower class for stationary Gaussian processes. Trans. Amer. Math. Soc., 147:301–331, 1970

  40. [48]

    D. Slepian. The one-sided barrier problem for Gaussian noise.Bell System Tech. J., 41:463– 501, 1962

  41. [49]

    Tindel, C

    S. Tindel, C. A. Tudor, and F. Viens. Sharp Gaussian regularity on the circle, and applications to the fractional stochastic heat equation.J. Funct. Anal., 217(2):280–313, 2004

  42. [50]

    C. A. Tudor and Y. Xiao. Sample paths of the solution to the fractional-colored stochastic heat equation.Stoch. Dyn., 17(1):1750004, 20, 2017

  43. [51]

    J. B. Walsh. An introduction to stochastic partial differential equations. In ´Ecole d’´ et´ e de probabilit´ es de Saint-Flour, XIV—1984, volume 1180 ofLecture Notes in Math., pages 265–

  44. [52]

    Y. Xiao. Strong local nondeterminism and sample path properties of Gaussian random fields. InAsymptotic theory in probability and statistics with applications, volume 2 ofAdv. Lect. Math. (ALM), pages 136–176. Int. Press, Somerville, MA, 2008

  45. [53]

    Y. Xiao. Sample path properties of anisotropic Gaussian random fields. InA minicourse on stochastic partial differential equations, volume 1962 ofLecture Notes in Math., pages 145–212. Springer, Berlin, 2009. School of Science and Engineering, The Chinese University of Hong Ko...

  46. [439]

    Springer, Berlin, 1986

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.