Paper: arXiv 2403.04674

Authors: David Itkin, Martin Larsson

Abstract

In the framework of stochastic portfolio theory we introduce rank volatility stabilized models for large equity markets over long time horizons. These models are rank-based extensions of the volatility stabilized models introduced by Fernholz & Karatzas in 2005. On the theoretical side we establish global existence of the model and ergodicity of the induced ranked market weights. We also derive explicit expressions for growth-optimal portfolios and show the existence of relative arbitrage with respect to the market portfolio. On the empirical side we calibrate the model to sixteen years of CRSP US equity data matching (i) rank-based volatilities, (ii) stock turnover as measured by market weight collisions, (iii) the average market rate of return and (iv) the capital distribution curve. Assessment of model fit and error analysis is conducted both in and out of sample. To the best of our knowledge this is the first model exhibiting relative arbitrage that has statistically been shown to have a good quantitative fit with the empirical features (i)-(iv). We additionally simulate trajectories of the calibrated model and compare them to historical trajectories, both in and out of sample.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper is dense with stochastic calculus, SDEs, and theoretical proofs (existence, ergodicity, arbitrage), placing it in the high math category. It also demonstrates high empirical rigor by calibrating to 16 years of CRSP data, performing in/out-of-sample error analysis, and simulating trajectories.

Research Flowchart

  flowchart TD
  A["Research Goal"] --> B["Data Source"]
  B --> C["Model Calibration"]
  C --> D["Computational Analysis"]
  D --> E["Key Findings/Outcomes"]

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 proposes rank volatility stabilized models for large equity markets over long horizons.

Source sentence
we introduce rank volatility stabilized models for large equity markets over long time horizons.

Assumes 4

  • Parameter condition (2.5) holds, ensuring global existence and strictly positive market capitalizations. model
    Source sentence
    we assume that the parameter condition (2.5) holds so that we have global existence for S and all assets have strictly positive market capitalizations.
  • The analysis is carried out on a filtered probability space satisfying the usual assumptions. other
    Source sentence
    we work on a filtered probability space (Ω, F, (F_t)_{t≥0}, P) satisfying the usual assumptions
  • Only common stock equities are included in the CRSP dataset. data
    Source sentence
    we filter equities to only include common stock
  • No trading frictions (transaction costs) are considered in the portfolio performance analysis. execution
    Source sentence
    we do not consider any trading frictions here.

Data

  • CRSP equity universe daily price data · CRSP market-prices public
    1990-01-01 to 2005-12-31 · US equities, at least 3500 stocks
    Source sentence
    We calibrate the model to US equity data for the 16-year period from Jan 1, 1990 to Dec 31, 2005. We use daily price data from the CRSP equity universe.
  • CRSP equity universe daily price data (out‑of‑sample) · CRSP market-prices public
    2006-01-01 to 2022-12-31 · US equities, same cleaning as in‑sample
    Source sentence
    In Section 4 we also assess the performance of the model out-of-sample using analogously processed data for the time period Jan 1, 2006 to Dec 31, 2022.

Recognised sources: CRSP

Costs

Unclear - sources disagree The platform classifier and the paper-text extraction give different answers, so no value is asserted.

Platform classifier: Not applicable

Paper text extraction: Not modeled

Source sentence (the authors' words)
we do not consider any trading frictions here.

Finds 5

  • The model admits a globally existing solution and the ranked market weights are ergodic. theoretical
    Source sentence
    On the theoretical side we establish global existence of the model and ergodicity of the induced ranked market weights.
  • Relative arbitrage with respect to the market portfolio exists in the rank‑volatility stabilized market. theoretical
    Source sentence
    show the existence of relative arbitrage with respect to the market portfolio.
  • A calibrated parameter λ≈0.11 simultaneously matches the historical market rate of return, collisions and capital‑distribution curve, providing the preferred model fit. pricing-calibration
    Source sentence
    value λ = 0.11 that matches the historical market rate of return also turns out to offer a balanced fit to both the empirical collisions and capital distribution curves.
  • Out‑of‑sample collision errors are larger than in‑sample errors, indicating some over‑fitting or estimation noise. statistical
    Source sentence
    normalized errors are larger out-of-sample than in-sample.
  • Simulated trajectories are not strongly dependent on the choice of λ, suggesting robustness of the synthetic data generation. other
    Source sentence
    the behaviour of individual trajectories is not strongly dependent on the choice of λ.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors): Macroscopic properties of equity markets: stylized facts and … (uses its method or data)

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.