pith. sign in

arxiv: 1907.06782 · v1 · pith:R2YI3LN2new · submitted 2019-07-15 · 🧮 math.PR

AR(1) processes driven by second-chaos white noise: Berry-Ess\'een bounds for quadratic variation and parameter estimation

Soukaina Douissi , Khalifa Es-Sebaiy , Fatimah Alshahrani , Frederi G. Viens This is my paper

Pith reviewed 2026-05-24 21:01 UTC · model grok-4.3

classification 🧮 math.PR
keywords AR(1) processesquadratic variationBerry-Esséen boundssecond Wiener chaosparameter estimationtotal variationmean-reversionWiener space analysis
0
0 comments X

The pith

AR(1) processes with second-chaos white noise have quadratic variation converging to normal at a bounded total-variation rate, which supports mean-reversion estimation.

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

The paper aims to establish an upper bound on the speed of convergence in total variation to the normal law for the quadratic variation of AR(1) processes driven by second-chaos white noise. It uses this bound to study the estimation of the mean-reversion parameter in the model. A sympathetic reader would care because such bounds provide quantitative rates for central limit theorems in non-standard settings, potentially improving statistical inference for time series with dependent or non-Gaussian noise.

Core claim

For AR(1) processes driven by white noise in the second Wiener chaos, the normalized quadratic variation converges in total variation to the standard normal distribution at a rate that can be bounded from above using tools from the analysis on Wiener space; this bound is then applied to derive properties of the estimator for the mean-reversion coefficient.

What carries the argument

The quadratic variation of the AR(1) process, whose normalized version's distance to normality is bounded via Wiener space analysis.

If this is right

  • The total variation distance between the law of the quadratic variation and the normal is upper bounded.
  • This bound applies directly to the asymptotic analysis of the mean-reversion estimator.
  • Simulations confirm the theoretical convergence rates for the quadratic variation and the estimator.
  • The results extend classical Berry-Esséen theorems to this class of processes with second-chaos driving noise.

Where Pith is reading between the lines

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

  • If the bound holds, similar quantitative CLTs might apply to quadratic variations in other linear stochastic processes with chaos noise.
  • The estimation procedure could be extended to test hypotheses about the mean-reversion parameter using the rate information.
  • Connections to Malliavin calculus suggest potential for deriving bounds in related models like continuous-time Ornstein-Uhlenbeck processes.

Load-bearing premise

The driving noise is white noise belonging to the second Wiener chaos.

What would settle it

Numerical computation showing that the total-variation distance to the normal law for the quadratic variation does not decrease according to the derived upper bound as the sample size increases.

Figures

Figures reproduced from arXiv: 1907.06782 by Fatimah Alshahrani, Frederi G. Viens, Khalifa Es-Sebaiy, Soukaina Douissi.

Figure 1
Figure 1. Figure 1: 500 various observations from (8) for different values of [PITH_FULL_IMAGE:figures/full_fig_p010_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Asymptotic variance for values of |a1| between 0.09 and 0.99 To investigate the asymptotic distribution of aˆn empirically, we need to compare the distri￾bution of the following statistic φ(n, a1) := p 2a 2 p 1 (1 − a 2 1 )(5 − 4a 2 1 ) √ n(ˆan − |a1|) (41) with the standard normal distribution N (0, 1). For this aim, for parameter choices |a1| = 0.5, n = 3000, σ = 1, and based on 3000 replications, we obt… view at source ↗
Figure 3
Figure 3. Figure 3: Histogram of φ(n, a1) with n = 3000, |a1| = 0.5, σ = 1, 3000 replications. This [PITH_FULL_IMAGE:figures/full_fig_p033_3.png] view at source ↗
read the original abstract

In this paper, we study the asymptotic behavior of the quadratic variation for the class of AR(1) processes driven by white noise in the second Wiener chaos. Using tools from the analysis on Wiener space, we give an upper bound for the total-variation speed of convergence to the normal law, which we apply to study the estimation of the model's mean-reversion. Simulations are performed to illustrate the theoretical results.

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

0 major / 2 minor

Summary. The paper studies the asymptotic behavior of the quadratic variation for AR(1) processes driven by white noise belonging to the second Wiener chaos. Using Malliavin calculus and Stein's method on Wiener space, it derives an upper bound on the total-variation distance to the normal law for the normalized quadratic variation. This bound is applied to obtain convergence rates for the estimator of the mean-reversion parameter, with numerical simulations provided to illustrate the results.

Significance. If the derivations are correct, the work extends Berry-Esseen-type results to a non-Gaussian setting via standard Wiener-space tools and directly links the bounds to statistical estimation rates. The explicit TV bound and its use for parameter estimation constitute a concrete contribution to the literature on limit theorems and inference for chaos-driven processes.

minor comments (2)
  1. [Abstract] Abstract: the claim of an 'upper bound' would benefit from a brief indication of its dependence on the model parameters (e.g., the AR coefficient or the chaos variance) to give readers an immediate sense of the result's sharpness.
  2. [Introduction] The manuscript would be strengthened by an explicit statement of the precise normalization used for the quadratic variation (e.g., centering and scaling constants) already in the introduction or statement of the main theorem.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the careful reading, positive summary, and recommendation of minor revision. The report lists no specific major comments, so we have no individual points to address.

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper derives an explicit upper bound on total-variation distance to normality for the quadratic variation of an AR(1) process driven by second-chaos white noise, using standard Malliavin calculus and Stein-method tools on Wiener space; this bound is then applied to obtain convergence rates for the mean-reversion estimator. The model is defined directly by the noise class, the derivation chain relies on external, non-self-referential analytic machinery rather than any fitted parameter renamed as a prediction or any load-bearing self-citation, and no equation reduces by construction to its own inputs. The abstract and setup contain no self-definitional steps, uniqueness theorems imported from the authors' prior work, or ansatz smuggling.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only the abstract is available, so the ledger is necessarily incomplete and limited to what is explicitly named.

axioms (1)
  • domain assumption The driving noise belongs to the second Wiener chaos and is white.
    Stated in the title and abstract as the model class under study.

pith-pipeline@v0.9.0 · 5612 in / 1179 out tokens · 25684 ms · 2026-05-24T21:01:02.443285+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

54 extracted references · 54 canonical work pages · 4 internal anchors

  1. [1]

    and Morlanes, G

    Azmoodeh, E. and Morlanes, G. I. (2013). Drift parameter estimation for fractional Ornstein- Uhlenbeck process of the second kind.Statistics. DOI: 10.1080/02331888.2013.863888. 34

    work page doi:10.1080/02331888.2013.863888 2013

  2. [2]

    and Viitasaari, L

    Azmoodeh, E. and Viitasaari, L. (2015). Parameter estimation based on discrete observations of fractional Ornstein-Uhlenbeck process of the second kind.Statist. Infer. Stoch. Proc.18, no. 3, 205-227

    work page 2015

  3. [3]

    Balakrishna, N.andShiji, K.(2014).ExtremeValueAutoregressiveModelandItsApplications, Journal of Statistical Theory and Practice, 8 (3), 460–481

    work page 2014

  4. [4]

    and Viens, F

    Barboza, L.A. and Viens, F. (2017). Parameter estimation of Gaussian stationary processes using the generalized method of moments.Electron. J. Statist.11 (1), 401-439

    work page 2017

  5. [5]

    and Ouknine, Y

    Belfadli, R., Es-Sebaiy, K. and Ouknine, Y. (2011). Parameter Estimation for Fractional Ornstein-Uhlenbeck Processes: Non-Ergodic Case.Frontiers in Science and Engineering (An International Journal Edited by Hassan II Academy of Science and Technology). 1, no. 1, 1-16

    work page 2011

  6. [6]

    and Peccati, G

    Biermé, H., Bonami, A., Nourdin, I. and Peccati, G. (2012). Optimal Berry-Esséen rates on the Wiener space: the barrier of third and fourth cumulants.ALEA 9, no. 2, 473-500

    work page 2012

  7. [7]

    and Iacus, S

    Brouste, A. and Iacus, S. M. (2012). Parameter estimation for the discretely observed fractional Ornstein-Uhlenbeck process and the Yuima R package.Comput. Stat.28, no. 4, 1529-1547

    work page 2012

  8. [8]

    and Maejima, M

    Cheridito, P., Kawaguchi, H. and Maejima, M. (2003). Fractional Ornstein-Uhlenbeck pro- cesses, Electr. J. Prob.8, 1-14

    work page 2003

  9. [9]

    L.H.Y. Chen, L. Goldstein and Q.-M. Shao (2011).Normal Approximation by Stein’s Method. Berlin: Springer-Verlag

    work page 2011

  10. [10]

    and Sykes, Z.M

    Cumberland, W.G. and Sykes, Z.M. (1982). Weak Convergence of an Autoregressive Process Used in Modeling Population Growth.Journal of Applied Probability, 19 (2) 450-455

    work page 1982

  11. [11]

    and Martynova, G

    Davydov, Y.A. and Martynova, G. V. (1987). Limit behavior of multiple stochastic integral. Statistics and control of random process.Preila, Nauka, Moscow, 55-57 (in Russian)

    work page 1987

  12. [12]

    Dobrushin, R. L. and Major, P. (1979). Non-central limit theorems for nonlinear functionals of Gaussian fields.Z. Wahrsch. verw. Gebiete, 50, 27-52

    work page 1979

  13. [13]

    Douissi, S., Es-Sebaiy, K., Viens, F. (2019). Berry-Esséen bounds for parameter estimation of general Gaussian processes.Latin American journal of probability and mathematical statistics, 16, 633-664

    work page 2019

  14. [14]

    and Ouknine, Y

    El Machkouri, M., Es-Sebaiy, K. and Ouknine, Y. (2015). Least squares estimator for non- ergodic OrnsteinUhlenbeck processes driven by Gaussian processes.Journal of the Korean Sta- tistical Society, DOI: 10.1016/j.jkss.2015.12.001 (In press)

    work page doi:10.1016/j.jkss.2015.12.001 2015

  15. [15]

    and Tudor, C

    El Onsy, B., Es-Sebaiy, K. and Tudor, C. (2014). Statistical analysis of the non-ergodic frac- tional Ornstein-Uhlenbeck process of the second kind.Preprint

    work page 2014

  16. [16]

    and Viens, F

    El Onsy, B., Es-Sebaiy, K. and Viens, F. (2017). Parameter Estimation for Ornstein-Uhlenbeck driven by fractional Ornstein-Uhlenbeck processes.Stochastics, Vol. 89, No. 2, 431-468. 35

    work page 2017

  17. [17]

    Ernst, Ph., Brown, L., Shepp, L., Wolpert, R. (2017). Stationary Gaussian Markov processes as limits of stationary autoregressive time series.Journal of Multivariate Analysis, 155, 180-186

    work page 2017

  18. [18]

    and Viens, F

    Es-Sebaiy, K. and Viens, F. (2018). Optimal rates for parameter estimation of stationary Gaus- sian processes.Stochastic Processes and their Applications, in press, 49 pages, available online DOI: 10.1016/j.spa.2018.08.010

    work page doi:10.1016/j.spa.2018.08.010 2018

  19. [19]

    Régularité des trajectoires des fonctions alé atoires gaussiennes

    Fernique, X. Régularité des trajectoires des fonctions alé atoires gaussiennes. InEcole d’Et´e de Probabilités de Saint-Flour IV-1974, Lecture Notes in Math.480, 1-96. Springer V., 1975

    work page 1974

  20. [20]

    (1996).A unified View of linear AR(1) models, preprint

    Grunwald, G.K., Hyndman, R.J., and Tedesco, L.M. (1996).A unified View of linear AR(1) models, preprint

    work page 1996

  21. [21]

    On Non-Gaussian AR(1) Inflation Modeling

    Hürlimann, W. (2012). “On Non-Gaussian AR(1) Inflation Modeling” Journal of Statistical and Econometric Methods,1 (1), 93-109

    work page 2012

  22. [22]

    and Nualart, D

    Hu, Y. and Nualart, D. (2010). Parameter estimation for fractional Ornstein-Uhlenbeck pro- cesses. Statist. Probab. Lett.80, 1030-1038

    work page 2010

  23. [23]

    and Song, J

    Hu, Y. and Song, J. (2013). Parameter estimation for fractional Ornstein-Uhlenbeck processes with discrete observations.F. Viens et al (eds), Malliavin Calculus and Stochastic Analysis: A Festschrift in Honor of David Nualart, 427-442, Springer

    work page 2013

  24. [24]

    and Le Breton, A

    Kleptsyna, M. and Le Breton, A. (2002). Statistical analysis of the fractional Ornstein- Uhlen- beck type process.Statist. Infer. Stoch. Proc.5, 229-241

    work page 2002

  25. [25]

    and Neuenkirch, A

    Kloeden, P. and Neuenkirch, A. (2007). The pathwise convergence of approximation schemes for stochastic differential equations.LMS J. Comp. Math.10, 235-253

    work page 2007

  26. [26]

    Liptser, R. S. and Shiryaev, A. N. (2001).Statistics of Random Processes: II. Applications, Springer-Verlag, New York

    work page 2001

  27. [27]

    Maddison, C.J., Tarlow, D., Minka, T. (2015). A* Sampling, Preprint, available online at https://arxiv.org/abs/1411.0030v2

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  28. [28]

    Nakajima, J., Kunihama, T., Omori, Y., and Frühwirth-Schnatter, S. (2012). Generalized extreme value distribution with time-dependence using the AR and MA models in state space form. Computational Statistics & Data Analysis, 56 (11), 3241-3259

    work page 2012

  29. [29]

    A third-moment theorem and precise asymptotics for variations of stationary Gaussian sequences

    Neufcourt, L. and Viens, F. (2014). A third-moment theorem and precise asymptotics for variations of stationary Gaussian sequences. In press in ALEA; preprint available at http://arxiv.org/abs/1603.00365

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  30. [30]

    and Tindel, S

    Neuenkirch, A. and Tindel, S. (2014). A least square-type procedure for parameter estimation in stochastic differential equations with additive fractional noise.Statist. Infer. Stoch. Proc.17, no. 1, 99-120. 36

    work page 2014

  31. [31]

    and Peccati, G

    Nourdin, I. and Peccati, G. (2015). The optimal fourth moment theorem.Proc. Amer. Math. Soc.143, 3123-3133

    work page 2015

  32. [32]

    and Peccati, G

    Nourdin, I. and Peccati, G. (2012). Normal approximations with Malliavin calculus : from Stein’s method to universality.Cambridge Tracts in Mathematics 192. Cambridge University Press, Cambridge

    work page 2012

  33. [33]

    Nourdin, I., Peccati, G., and Yang, X. (2018). Berry-Esseen bounds in the Breuer-Major CLT and Gebelein’s inequality. Preprint https://arxiv.org/abs/1812.08838

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  34. [34]

    Novikov, A.A. (1971). Sequential estimation of the parameters of diffusion type processes.Teor. Veroyatn. Primen., 16 (2), 394-396

    work page 1971

  35. [36]

    and Peccati, G

    Nualart, D. and Peccati, G. (2005). Central limit theorems for sequences of multiple stochastic integrals. Ann. Probab. 33, no. 1, 177-193

    work page 2005

  36. [37]

    Total variation estimates in the Breuer-Major theorem

    D. Nualart and H. Zhou (2018): Total variation estimates in the Breuer-Major theorem. Preprint arXiv:1807.09707

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  37. [38]

    Supremum concentration inequality and modulus of continuity for sub- nth chaos processes.J

    Viens, F.; Vizcarra, A. Supremum concentration inequality and modulus of continuity for sub- nth chaos processes.J. Functional Analysis248 (1) (2007), 1-26

    work page 2007

  38. [39]

    Nualart, D. (2006). The Malliavin calculus and related topics.Springer-Verlag, Berlin

    work page 2006

  39. [40]

    Üstünel, A. S. (1995). An Introduction to Analysis on Wiener Space. Lecture Notes in Math

    work page 1995

  40. [41]

    Appl.118 614- 628

    Nualart, D.andOrtiz-Latorre, S.(2008).Centrallimittheoremsformultiplestochasticintegrals and Malliavin calculus.Stochastic Process. Appl.118 614- 628

    work page 2008

  41. [42]

    and Peccati, G

    Nualart, D. and Peccati, G. (2005). Central limit theorems for sequences of multiple stochastic integrals. Ann. Probab.33 177-193

    work page 2005

  42. [43]

    and Tudor, C

    Peccati, G. and Tudor, C. A. (2004). Gaussian limits for vector-valued multiple stochastic inte- grals. In Séminaire de Probabilités XXXVIII. Lecture Notes in Math. 1857 247-262.Springer, Berlin

    work page 2004

  43. [44]

    Schoenberg, I.J. (1951). On Polya frequency functions.Journal d’Analyse Mathématique, 1, 331-374

    work page 1951

  44. [45]

    Shepp, L.A. (1969). Explicit solutions to some problems of optimal stopping.The Annals of Mathematical Statistics40, 993-1010

    work page 1969

  45. [46]

    and Shiryaev, A.N

    Shepp, L.A. and Shiryaev, A.N. (1993). The Russian option: reduced regret.The Annals of Applied Probability3, 631-640. 37

    work page 1993

  46. [47]

    and Vardi, Y

    Shepp, L.A. and Vardi, Y. (1982) Maximum likelihood reconstruction for emission tomography. IEEE transactions on Medical Imaging, 1, 113-122

    work page 1982

  47. [48]

    (2017).Time Series Analysis: Nonstationary and Noninvertible Distribution The- ory, volume 16

    Tanaka, K. (2017).Time Series Analysis: Nonstationary and Noninvertible Distribution The- ory, volume 16. Wiley New York

    work page 2017

  48. [49]

    Gwladys Toulemonde, Armelle Guillou, Philippe Naveau, Mathieu Vrac and Frederic Cheval- lier. (2010). Autoregressive models for maxima and their applications to CH4 and N2O. Envi- ronmetrics, 21 (2), 189-207

    work page 2010

  49. [50]

    Tudor, C.A. (2008). Analysis of the Rosenblatt process.ESAIM: Probability and Statistics, 12, 230–257

    work page 2008

  50. [51]

    and Viens, F

    Tudor, C. and Viens, F. (2007). Statistical aspects of the fractional stochastic calculus.Ann. Statist. 35, no. 3, 1183-1212

    work page 2007

  51. [52]

    Representation of Stationary and stationary increment processes via Langevin equation and self-similar processes

    Viitasaari, L. Representation of Stationary and stationary increment processes via Langevin equation and self-similar processes. (2016).Statistics and Probability Letters, 115, 45-53

    work page 2016

  52. [53]

    Voutilainen, M., Viitasaari, L., and Ilmonen, P. (2017). On model fitting and estimation of strictly stationary processes.Modern Stochastics : Theory and Applications. 4 (4), 381–406

    work page 2017

  53. [54]

    Voutilainen, M., Viitasaari, L., and Ilmonen, P. (2019). Note on AR(1) characterization of sta- tionary processes and model fitting.Modern Stochastics : Theory and Applications, to appear, https://doi.org/10.15559/19-VMSTA132

    work page doi:10.15559/19-vmsta132 2019

  54. [55]

    Yuan Yan, Genton, M.C. (2018). Non-Gaussian autoregressive processes with Tukey g-and-h transformations. Environmetrics, e2503, https://doi.org/10.1002/env.2503 . 38

    work page doi:10.1002/env.2503 2018