Paper: arXiv 2511.13277

Authors: Jutta G. Kurth, Jean-Philippe Bouchaud

Abstract

We derive the stationary distribution in various regimes of the extended Chiarella model of financial markets. This model is a stochastic nonlinear dynamical system that encompasses dynamical competition between a (saturating) trending and a mean-reverting component. We find the so-called mispricing distribution and the trend distribution to be unimodal Gaussians in the small noise, small feedback limit. Slow trends yield Gaussian-cosh mispricing distributions that allow for a P-bifurcation: unimodality occurs when mean-reversion is fast, bimodality when it is slow. The critical point of this bifurcation is established and refutes previous ad-hoc reports and differs from the bifurcation condition of the dynamical system itself. For fast, weakly coupled trends, deploying the Furutsu-Novikov theorem reveals that the result is again unimodal Gaussian. For the same case with higher coupling we disprove another claim from the literature: bimodal trend distributions do not generally imply bimodal mispricing distributions. The latter becomes bimodal only for stronger trend feedback. The exact solution in this last regime remains unfortunately beyond our proficiency.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper is highly theoretical, deriving analytical solutions for stationary distributions using advanced stochastic calculus (Fokker-Planck, Lyapunov equations, Furutsu-Novikov theorem) and bifurcation analysis. It lacks any empirical validation, backtests, or data-driven implementation details, focusing solely on mathematical proofs and limits.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Determine Stationary Distributions<br>of Mode-switching Chiarella Model"] --> B["Methodology:<br>Analytical Derivation &<br>Furutsu-Novikov Theorem"]
  B --> C{"Parameter Regimes"}
  C --> D["Small Noise,<br>Small Feedback"]
  C --> E["Slow Trends,<br>Strong Feedback"]
  C --> F["Fast, Weakly Coupled<br>Trends"]
  D --> G["Outcomes:<br>Unimodal Gaussians<br>Mispricing & Trend"]
  E --> H["Outcomes:<br>Gaussian-Cosh Mispricing<br>P-bifurcation Point"]
  F --> I["Outcomes:<br>Unimodal Gaussian Trend<br>Disproves Bimodality Claim"]
  H --> J["Key Finding:<br>Crit. Point Refutes<br>Previous Ad-hoc Reports"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper derives stationary distributions of mispricing and trend signals in various regimes of the extended Chiarella model of financial markets.

Source sentence
We derive the stationary distribution in various regimes of the extended Chiarella model of financial markets. This model is a stochastic nonlinear dynamical system that encompasses dynamical competition between a (saturating) trending and a mean-reverting component.

Assumes 3

  • The analysis assumes the small‑noise, small‑feedback limit (γσ_M ≪ 1). model
    Source sentence
    As anticipated above, these re-sults hold provided the condition γσM ≪1 is satisfied.
  • The drift matrix A must have non‑positive eigenvalues so that the system does not diverge and a stationary distribution exists. model
    Source sentence
    Sec. II). This is always true in the considered limit. If it was not true, the drift matrix A would have positive eigenvalues, i.e. the system would diverge and no stationary distribution would exist.
  • When the trend time scale α⁻¹ is short compared to the mean‑reversion time κ⁻¹ and the positive feedback β is not very strong, the Furutsu‑Novikov theorem can be applied. model
    Source sentence
    Third, a stationary distribution in the case where the trend time scale α−1 is short compared to the typical mean-reversion time κ−1, while the positive feedback term β is not very strong, is found via the Furutsu-Novikov theorem. In this case the distribution is uni-modal and Gaussian, too.

Data

No dataset named in the text.

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • In the small‑noise, small‑feedback limit the mispricing distribution is Gaussian and unimodal. theoretical
    Source sentence
    We find the so-called mispricing distribution and the trend distribution to be unimodal Gaussians in the small noise, small feedback limit. From the joint probability distribution p(δ, M), the mispricing distri-bution p(δ) can be obtained through marginalisation, p(δ) = Z ∞ −∞ p(δ, M) dM = 1 p 2πσ2 δ e −δ2 2σ2 δ , (13) such that in this limit the mispricing distribution is Gaussian and thus unimodal.
  • Slow trends produce Gaussian‑cosh mispricing distributions that exhibit a P‑bifurcation: unimodal when mean‑reversion is fast, bimodal when it is slow. theoretical
    Source sentence
    Slow trends yield Gaussian-cosh mispricing distributions that allow for a P-bifurcation: unimodality occurs when mean-reversion is fast, bimodality when it is slow.
  • For fast, weakly coupled trends the stationary distribution remains unimodal Gaussian, as shown via the Furutsu‑Novikov theorem. theoretical
    Source sentence
    For fast, weakly coupled trends, deploying the Furutsu-Novikov theorem reveals that the result is again unimodal Gaussian. In this case the distribution is uni-modal and Gaussian, too.
  • Bimodal trend distributions do not generally imply bimodal mispricing distributions; both become bimodal only when the feedback parameter β is sufficiently strong. theoretical
    Source sentence
    For the same case with higher coupling we disprove another claim from the literature: bimodal trend distributions do not generally imply bimodal mispricing distributions. If, in turn, the positive feed-back is strong, we find – disproving another common claim in the literature – that bimodality in the trend distribution does not necessarily imply bimodality in the mispricing distribution when the price variable has
  • The exact analytical solution for the stationary distribution in the strong‑feedback regime is not obtained. negative
    Source sentence
    The exact solution in this last regime remains unfortunately beyond our proficiency. We have unfortunately not been able to find an ex-act analytical solution for the stationary distribution in these last cases, for reasons that we explain in the text and in the corresponding Appendix.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 15 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (83)

  1. Stochastic processes in physics and chemistry (1992)
  2. Reaction-rate theory: fifty years after Kramers (1990)
  3. Nonlinear oscillations, dynamical systems, and bifurcations of vector fields (2013)
  4. Nonlinear dynamical economics and chaotic motion (1993)
  5. Can heterogeneous agent models explain the alleged mispricing of the S&P 500? (2021)
  6. Market force, ecology and evolution (2002)
  7. Bayesian filtering and smoothing (2023)
  8. A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains (1970)
and 75 more
  1. Maximum likelihood from incomplete data via the EM algorithm (1977)
  2. The EM approach to the multiple indicators and multiple causes model via the estimation of the latent variable (1981)
  3. On The Dynamic Behavior of Prices in Disequilibrium (1980) — link
  4. Efficient capital markets (1970)
  5. Efficient capital markets: II (1991)
  6. Do Stock Prices Move Too Much to Be Justified by Subsequent Changes in Dividends? (1981)
  7. On the behaviour of commodity prices (1992)
  8. Empirical exchange rate models of the seventies: Do they fit out of sample? (1983)
  9. Yield spreads and interest rate movements: A bird’s eye view (1991)
  10. Chapter 15 Anomalies and market efficiency (2003) — link
  11. Value and momentum everywhere (2013)
  12. Some anomalous evidence regarding market efficiency (1978)
  13. Security Analysis. New York: Whittlesey House (1934)
  14. Does the stock market overreact? (1985)
  15. Persuasive evidence of market inefficiency (1985)
  16. Does the stock market rationally reflect fundamental values? (1986)
  17. Stock prices, earnings, and expected dividends (1988)
  18. The cross-section of expected stock returns (1992)
  19. Size, value, and momentum in international stock returns (2012)
  20. Contrarian investment, extrapolation, and risk (1994)
  21. Black was right: Price is within a factor 2 of Value (2018) — link
  22. Evidence of predictable behavior of security returns (1990)
  23. Returns to buying winners and selling losers: Implications for stock market efficiency (1993)
  24. Fads, martingales, and market efficiency (1990)
  25. On persistence in mutual fund performance (1997)
  26. Time series momentum (2012)
  27. The excess volatility puzzle explained by financial noise amplification from endogenous feedbacks (2022)
  28. Two centuries of trend following (2014) — link
  29. Investor behavior in the October 1987 stock market crash: Survey evidence (1987)
  30. Irrational Exuberance, Princeton Univ (2000)
  31. Excess volatility (2006)
  32. Advances in Behavioral Finance (Russell Sage Foundation, New York) (1993)
  33. Advances in behavioral finance, Volume II (2005)
  34. Scaling and criticality in a stochastic multi-agent model of a financial market (1999)
  35. Exploration of the Parameter Space in Macroeconomic Models (2024)
  36. How market ecology explains market malfunction (2021)
  37. Information geometry for multiparameter models: New perspectives on the origin of simplicity (2022)
  38. A century of evidence on trend-following investing (2017) — link
  39. The inelastic market hypothesis: a microstructural interpretation (2022)
  40. Noise (1986)
  41. Ponzi funds — → 2405.12768
  42. In Search of the Origins of Financial Fluctuations: The Inelastic Markets Hypothesis (2021)
  43. Propagation of a higher-order cosh-Gaussian beam in turbulent atmosphere (2011)
  44. Wave propagation and scattering in random media (1978)
  45. Trades, quotes and prices: financial markets under the microscope (2018)
  46. Investor psychology and security market under-and overreactions (1998)
  47. Fast and slow thinking (2011)
  48. New experimental evidence on expectations formation (2017)
  49. Heterogeneous agent models in economics and finance (2006)
  50. Agent-based computational finance (2006)
  51. Heterogeneity, market mechanisms, and asset price dynamics (2009)
  52. Heterogeneous agent models in finance (2018)
  53. The dynamics of speculative behaviour (1992)
  54. An analysis of the effect of noise in a heterogeneous agent financial market model (2011)
  55. Co-existence of trend and value in financial markets: Estimating an extended Chiarella model (2020) — link
  56. Revisiting the Excess Volatility Puzzle Through the Lens of the Chiarella Model — → 2505.07820
  57. Universally sloppy parameter sensitivities in systems biology models (2007)
  58. Quadratic Hawkes processes for financial prices (2017)
  59. Zooming in on equity factor crowding (2020) — link
  60. The Footprint of Trend-Following: Can CTAs really move the market? (2022) — link
  61. Continuous auctions and insider trading (1985)
  62. Behavioral heterogeneity in stock prices (2007)
  63. The socio-economic dynamics of speculative markets: interacting agents, chaos, and the fat tails of return distributions (1998)
  64. Nonlinear dynamics and chaos: with applications to physics, biology, chemistry, and engineering (2018)
  65. How market ecology explains market malfunction (2021)
  66. On the calibration of Silverman’s test for multimodality (2001)
  67. Dynamic portfolio choice with frictions (2016)
  68. On the bimodality of the distribution of the S&P 500’s distortion: Empirical evidence and theoretical explanations (2017)
  69. Using kernel density estimates to investigate multimodality (1981)
  70. Trends, reversion, and critical phenomena in financial markets (2021)
  71. Trends’ signal strength and the performance of CTAs (2019)
  72. The stochastic bifurcation behaviour of speculative financial markets (2008)
  73. CRSP US Stock Database: Daily Stock Prices (2025)
  74. A new metric for probability distributions (2003)
  75. Some asymptotics for multimodality tests based on kernel density estimates (1992)

2 of 83 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.