pith. sign in

arxiv: 2605.22933 · v1 · pith:JSBGQNZ3new · submitted 2026-05-21 · ❄️ cond-mat.stat-mech · physics.class-ph· physics.comp-ph· physics.data-an· physics.pop-ph

Emergent heavy-tailed distributions from a Markovian random walk

Henrique S. Lima , Evaldo M. F. Curado This is my paper

Pith reviewed 2026-05-25 05:42 UTC · model grok-4.3

classification ❄️ cond-mat.stat-mech physics.class-phphysics.comp-phphysics.data-anphysics.pop-ph
keywords random walkheavy-tailed distributionspower-law tailsMarkovian processstationary distributionOnsager-Machlup formalismscale-free statistics
0
0 comments X

The pith

A Markovian random walk with position-dependent steps yields a stationary Lorentz-like distribution with |x|^{-2} tails.

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

The paper demonstrates that heavy-tailed statistics can emerge from a strictly local, discrete-time Markovian random walk in one dimension. The step length is set by a deterministic function of the current position, which creates a positive feedback loop and induces effective correlations along trajectories. In the continuum limit this rule produces an exact closed-form stationary distribution proportional to (|x|/l + r Δx)^{-2}, with asymptotic power-law tails that decay as |x|^{-2} and remain normalizable. The non-zero initial acceleration is required both to drive the walker away from the origin and to ensure the scale-free form appears.

Core claim

The step length is governed by a deterministic function of the walker's position that establishes a positive feedback loop. Analytical derivations in the continuum limit together with numerical simulations yield the exact stationary distribution ρ_st(x) ∝ (|x|/l + r Δx)^{-2}, which exhibits power-law tails decaying as |x|^{-2} over six decades. Application of the Onsager-Machlup path-integral formalism shows that effective velocity and acceleration acquire physical meaning along shortest fluctuation trajectories, while a non-zero initial acceleration is the mechanism that both generates the scale-free statistics and guarantees normalizability of the distribution.

What carries the argument

The deterministic step-length function of position that establishes a positive feedback loop and effective correlations in the Markovian trajectories.

If this is right

  • The stationary distribution is robustly non-Gaussian and exhibits scale-free tails.
  • Effective velocity and acceleration acquire physical meaning along the shortest fluctuation trajectories.
  • The -2 power law can arise from a minimal local Markovian mechanism without memory or non-local jumps.
  • The distribution remains normalizable precisely because of the initial acceleration condition.

Where Pith is reading between the lines

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

  • This mechanism offers a candidate microscopic origin for the -2 power law seen in many complex systems, suggesting that some observed heavy tails may trace to local position feedback rather than long-range interactions or non-Markovian effects.
  • The same feedback construction could be tested in higher dimensions or with alternative deterministic functions to determine how generic the |x|^{-2} exponent remains.
  • The path-integral treatment of effective acceleration may connect the model to other descriptions of anomalous diffusion that rely on effective forces.

Load-bearing premise

The step length must be governed by a deterministic function of the walker's position that establishes a positive feedback loop, together with a non-zero initial acceleration.

What would settle it

Numerical realization of the exact step rule with zero initial acceleration that produces either a Gaussian stationary state or a non-normalizable distribution instead of the claimed Lorentz-like form with |x|^{-2} tails.

Figures

Figures reproduced from arXiv: 2605.22933 by Evaldo M. F. Curado, Henrique S. Lima.

Figure 1
Figure 1. Figure 1: Top left and right: ρ versus x for r = 1 and l = 10 on log-linear and log-log scales, respectively. Center left and right: ρ versus x for r = 2 and l = 10 on log-linear and log-log scales, respectively. Bottom left and right: ρ versus x for r = 1 and l = 100 on log-linear and log-log scales, respectively. 5 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
read the original abstract

The emergence of heavy-tailed statistics in complex systems is conventionally attributed to non-local stochastic jumps or non-Markovian memory. Here, we present a one-dimensional random walk where power-law behaviors arise instead from a strictly local, discrete-time Markovian mechanism. The step length is governed by a deterministic function of the walker's position, establishing a positive feedback loop that induces strong effective correlations along the trajectories. Through analytical derivations in the continuum limit and extensive numerical simulations, we show that this rule yields a robust, non-Gaussian stationary state. The exact analytical solution is obtained in the closed form of a symmetric, Lorentz-like distribution, $\rho_{\text{st}}(x) \propto (|x|/l+r\Delta x)^{-2}$, confirming asymptotic power-law tails that decay as $|x|^{-2}$ over six decades. Furthermore, by employing the Onsager-Machlup path-integral formalism, we demonstrate that effective velocity and acceleration acquire physical meaning along the shortest fluctuation trajectories. Crucially, we find that a non-zero initial acceleration acts as the fundamental mechanism driving the walker away from the origin, ensuring both the emergence of scale-free statistics and the normalizability of the stationary distribution. This minimal pathway provides a new microscopic foundation for the widespread $-2$ power law observed across multidisciplinary complex systems.

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 / 1 minor

Summary. The paper claims that a strictly local, discrete-time Markovian random walk with position-dependent deterministic step length (positive feedback) produces an exact closed-form stationary density ρ_st(x) ∝ (|x|/l + r Δx)^{-2} in the continuum limit, with |x|^{-2} power-law tails confirmed numerically over six decades; non-zero initial acceleration is required for both scale-free statistics and normalizability, and Onsager-Machlup formalism is used to interpret effective velocity and acceleration along shortest paths.

Significance. If the derivation is sound, the result supplies a minimal Markovian mechanism for the widespread -2 power law without invoking non-local jumps or memory, with the numerical confirmation over six decades providing concrete support. The use of the Onsager-Machlup path integral to assign meaning to effective velocity and acceleration is a methodological strength.

major comments (2)
  1. [Abstract] Abstract (and model definition): The claimed exact continuum-limit solution is given as ρ_st(x) ∝ (|x|/l + r Δx)^{-2}. This expression retains explicit dependence on the lattice spacing Δx. Taking the strict continuum limit Δx → 0 yields ρ_st(x) ∝ |x|^{-2}, whose integral ∫ dx/|x|^2 diverges at x=0 and is therefore non-normalizable. The retained Δx term indicates that the derivation does not fully eliminate the discretization scale, directly undermining the headline assertion of an exact, scale-free stationary state arising from the Markovian rule alone.
  2. [Abstract] Abstract (final paragraph) and model setup: Normalizability and the emergence of scale-free statistics are stated to require a non-zero initial acceleration, introduced as part of the model rather than derived from the Markovian rule. This additional assumption is load-bearing for both the claimed stationary distribution and its normalizability, yet it is presented as a 'crucial' mechanism without an explicit derivation showing how it follows from the position-dependent step length alone.
minor comments (1)
  1. The numerical confirmation is reported over six decades, but no error analysis, fitting procedure, or comparison against the exact functional form (including the Δx term) is described in the provided abstract.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on the continuum limit and the role of initial acceleration. We address each point below with clarifications and proposed revisions.

read point-by-point responses

  1. Referee: [Abstract] Abstract (and model definition): The claimed exact continuum-limit solution is given as ρ_st(x) ∝ (|x|/l + r Δx)^{-2}. This expression retains explicit dependence on the lattice spacing Δx. Taking the strict continuum limit Δx → 0 yields ρ_st(x) ∝ |x|^{-2}, whose integral ∫ dx/|x|^2 diverges at x=0 and is therefore non-normalizable. The retained Δx term indicates that the derivation does not fully eliminate the discretization scale, directly undermining the headline assertion of an exact, scale-free stationary state arising from the Markovian rule alone.

    Authors: We agree that the retained Δx term is important to note. The expression ρ_st(x) ∝ (|x|/l + r Δx)^{-2} is the exact stationary solution derived in the continuum approximation to the underlying discrete Markovian process; Δx functions as a small but finite regularization scale that ensures normalizability near the origin while the model remains fundamentally discrete. The strict Δx → 0 limit is singular, but the |x|^{-2} asymptotic tails hold for |x| ≫ r Δx and are robustly confirmed numerically over six decades. This does not contradict the Markovian origin of the power law. We will revise the abstract and model-definition section to explicitly state that the continuum limit is taken with Δx retained for normalizability. revision: partial

  2. Referee: [Abstract] Abstract (final paragraph) and model setup: Normalizability and the emergence of scale-free statistics are stated to require a non-zero initial acceleration, introduced as part of the model rather than derived from the Markovian rule. This additional assumption is load-bearing for both the claimed stationary distribution and its normalizability, yet it is presented as a 'crucial' mechanism without an explicit derivation showing how it follows from the position-dependent step length alone.

    Authors: The non-zero initial acceleration is identified through both analytic and numerical analysis as the mechanism that drives the walker away from the origin under the position-dependent step-length rule, thereby enabling the scale-free tails and normalizability. It arises directly from the positive-feedback dynamics when the deterministic step length is applied to initial conditions with non-zero velocity. We will add an explicit derivation in the model-setup section showing how this initial acceleration is induced by the Markovian position-dependent rule, making the connection clearer. revision: yes

Circularity Check

0 steps flagged

No circularity: stationary density is derived output, not input by construction.

full rationale

The paper defines a discrete-time Markovian rule with position-dependent deterministic step length (positive feedback) plus non-zero initial acceleration as modeling choices. It then states that analytical continuum-limit derivations yield the closed-form ρ_st(x) ∝ (|x|/l + r Δx)^{-2}. This is an output of solving the master equation or equivalent, not a self-definition or fitted parameter renamed as prediction. No self-citation chain, uniqueness theorem, or ansatz smuggling is invoked in the provided text to justify the central result. The retained Δx term and normalizability condition are model features analyzed for consistency, not reductions of the claimed derivation to its inputs. The derivation chain remains independent of the target distribution.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 1 invented entities

The central claim rests on postulating a deterministic position-dependent step-length rule and identifying non-zero initial acceleration as essential; these are introduced by model definition rather than derived from prior principles. The parameters appearing in the closed-form distribution are part of the solution but tied to the discretization and feedback strength.

free parameters (1)
  • l, r, Δx
    Scale and discretization parameters appearing in the exact stationary distribution form; their values are set by the choice of the deterministic step function and lattice spacing.
axioms (2)
  • domain assumption Step length is a deterministic function of current position that creates positive feedback
    Invoked in the model definition to generate effective correlations while remaining Markovian.
  • ad hoc to paper Non-zero initial acceleration is required for scale-free statistics and normalizability
    Identified as the fundamental mechanism in the final analytical step.
invented entities (1)
  • effective velocity and acceleration along shortest fluctuation trajectories no independent evidence
    purpose: To assign physical meaning via the Onsager-Machlup formalism
    Introduced to interpret the path-integral analysis; no independent falsifiable prediction is stated.

pith-pipeline@v0.9.0 · 5777 in / 1562 out tokens · 33267 ms · 2026-05-25T05:42:55.400856+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

40 extracted references · 40 canonical work pages

  1. [1]

    A. Einstein, ¨Uber die von der molekularkinetischen Theorie der W¨ arme geforderte Bewegung von in ruhenden Fl¨ ussigkeiten suspendierten Teilchen, Annalen der Physik322(8), 549–560 (1905)

    work page 1905

  2. [2]

    Chandrasekhar,Stochastic problems in physics and astronomy, Rev

    S. Chandrasekhar,Stochastic problems in physics and astronomy, Rev. Mod. Phys.15(1), 1–89 (1943). 7

    work page 1943

  3. [3]

    B. B. Mandelbrot,The variation of certain speculative prices, J. Business36(4), 394–419 (1963)

    work page 1963

  4. [4]

    R. N. Mantegna and H. E. Stanley,Scaling behaviour in the dynamics of an Italian market index, Nature 376(6535), 46–49 (1995)

    work page 1995

  5. [5]

    G. M. Viswanathan, S. V. Buldyrev, S. Havlin, M. G. E. da Luz, E. P. Raposo, and H. E. Stanley, Optimizing the success of random searches, Nature401(6756), 911–914 (1999)

    work page 1999

  6. [6]

    F. Xia, J. Liu, H. Nie, Y. Fu, L. Wan, and X. Kong, Random Walks: A Review of Algorithms and Applications, IEEE Trans. Emerg. Topics Comput. Intell.4, 95 (2020)

    work page 2020

  7. [7]

    Pearson,The Problem of the Random Walk, Nature72(1865), 294 (1905)

    K. Pearson,The Problem of the Random Walk, Nature72(1865), 294 (1905)

    work page 1905

  8. [8]

    Spitzer,Principles of Random Walk, 2nd ed

    F. Spitzer,Principles of Random Walk, 2nd ed. (Springer, 2001)

    work page 2001

  9. [9]

    Feller,An Introduction to Probability Theory and Its Applications, Vol

    W. Feller,An Introduction to Probability Theory and Its Applications, Vol. 1, 3rd ed. (Wiley, 1968)

    work page 1968

  10. [10]

    Bouchaud and A

    J.-P. Bouchaud and A. Georges,Anomalous diffusion in disordered media: statistical mechanisms, mod- els and physical applications, Phys. Rep.195(4-5), 127–293 (1990)

    work page 1990

  11. [11]

    Metzler and J

    R. Metzler and J. Klafter,The random walk’s guide to anomalous diffusion: a fractional dynamics approach, Phys. Rep.339(1), 1–77 (2000)

    work page 2000

  12. [12]

    M. F. Shlesinger, G. M. Zaslavsky, and J. Klafter,Strange kinetics, Nature363(6424), 31–37 (1993)

    work page 1993

  13. [13]

    Beck and C

    C. Beck and C. Tsallis,Anomalous velocity distributions in slow quantum-tunneling chemical reactions, Phys. Rev. Research7, L012081 (2025)

    work page 2025

  14. [14]

    E. W. Montroll and G. H. Weiss,Random walks on lattices. II, J. Math. Phys.6(2), 167–220 (1965)

    work page 1965

  15. [15]

    M. F. Shlesinger and J. Klafter,L´ evy Walks with Applications to Turbulent Diffusion, Phys. Rev. Lett. 54(23), 2551–2554 (1985)

    work page 1985

  16. [16]

    M. F. Shlesinger, G. M. Zaslavsky, and U. Frisch (Eds.),L´ evy Flights and Related Topics in Physics (Springer, Berlin, 1995)

    work page 1995

  17. [17]

    Zaburdaev, S

    V. Zaburdaev, S. Denisov, and J. Klafter,L´ evy walks, Rev. Mod. Phys.87(2), 483–530 (2015)

    work page 2015

  18. [18]

    Tirnakli, H

    U. Tirnakli, H. J. Jensen, and C. Tsallis,Restricted random walk model as a new testing ground for the applicability ofq-statistics, EPL96, 40008 (2011)

    work page 2011

  19. [19]

    Hanel and S

    R. Hanel and S. Thurner,Generalized(c, d)-entropy and aging random walks, Entropy15, 5324 (2013)

    work page 2013

  20. [20]

    M. E. J. Newman,Power laws, Pareto distributions and Zipf’s law, Contemp. Phys.46(5), 323–351 (2005)

    work page 2005

  21. [21]

    Clauset, C

    A. Clauset, C. R. Shalizi, and M. E. J. Newman,Power-law distributions in empirical data, SIAM Rev. 51(4), 661–703 (2009)

    work page 2009

  22. [22]

    Sornette,Critical Phenomena in Natural Sciences: Chaos, Fractals, Self-organization and Disorder: Concepts and Tools(Springer Science & Business Media, 2006)

    D. Sornette,Critical Phenomena in Natural Sciences: Chaos, Fractals, Self-organization and Disorder: Concepts and Tools(Springer Science & Business Media, 2006)

    work page 2006

  23. [23]

    B. B. Mandelbrot,The Fractal Geometry of Nature(W. H. Freeman, 1982)

    work page 1982

  24. [24]

    B. D. Hughes,Random Walks and Random Environments(Oxford University Press, 1995)

    work page 1995

  25. [25]

    Havlin and D

    S. Havlin and D. Ben-Avraham,Diffusion and reactions in fractals and disordered systems, Adv. Phys. 51(1), 187–292 (2002)

    work page 2002

  26. [26]

    B´ enichou and R

    O. B´ enichou and R. Voituriez,First-passage-time distributions in confined media, Phys. Rep.539(4), 225–284 (2014). 8

    work page 2014

  27. [27]

    Bouchaud and M

    J.-P. Bouchaud and M. Potters,Theory of Financial Risk and Derivative Pricing: From Statistical Physics to Risk Management(Cambridge University Press, 2003)

    work page 2003

  28. [28]

    V. A. Shneidman,Lognormal Size Distributions in Particle Growth Processes without Coagulation, Phys. Rev. Lett.80, 2386 (1998)

    work page 1998

  29. [29]

    Onsager and S

    L. Onsager and S. Machlup,Fluctuations and irreversible processes, Phys. Rev.91(6), 1505-1512 (1953)

    work page 1953

  30. [30]

    Machlup and L

    S. Machlup and L. Onsager,Fluctuations and irreversible processes. II, Phys. Rev.91(6), 1512-1515 (1953)

    work page 1953

  31. [31]

    P. V. Paraguass´ u and W. A. M. Morgado,Heat distribution of relativistic Brownian motion, Eur. Phys. J. B94, 197 (2021)

    work page 2021

  32. [32]

    Dr˘ agulescu and V

    A. Dr˘ agulescu and V. M. Yakovenko,Exponential and power-law probability distributions of wealth and income in the United Kingdom and the United States, Physica A299, 213 (2001)

    work page 2001

  33. [33]

    M. J. Aschwanden,Probability distribution functions of solar and stellar flares: I. Solar flare data, Physics4, 383 (2022)

    work page 2022

  34. [34]

    M. E. J. Newman,Self-organized criticality, evolution and the fossil extinction record, Proc. R. Soc. Lond. B263, 1605 (1996)

    work page 1996

  35. [35]

    Pradhan, A

    S. Pradhan, A. Hansen, and P. C. Hemmer,Crossover behavior in burst avalanches: Signature of imminent failure, Phys. Rev. E74, 016122 (2006)

    work page 2006

  36. [36]

    Camacho and R

    J. Camacho and R. V. Sol´ e,Scaling in ecological size spectra, Europhys. Lett.55, 774 (2001)

    work page 2001

  37. [37]

    Ferrer i Cancho and R

    R. Ferrer i Cancho and R. V. Sol´ e,Two regimes in the frequency of words and the origins of complex lexicons: Zipf’s law revisited, J. Quant. Linguist.8, 165 (2001)

    work page 2001

  38. [38]

    Barab´ asi and R

    A.-L. Barab´ asi and R. Albert,Emergence of scaling in random networks, Science286, 509 (1999)

    work page 1999

  39. [39]

    L. A. Adamic and B. A. Huberman,Power-law distribution of the World Wide Web, Science287, 2115a (2000)

    work page 2000

  40. [40]

    M. ´A. Serrano, D. Krioukov, and M. Bogu˜ n´ a,Self-similarity of complex networks, Nature489, 537 (2012). 9

    work page 2012