Paper: arXiv 2410.06568

Authors: Y. -F. Li, G. Papanicolaou

Abstract

Equity market dynamics are conventionally investigated in name space where stocks are indexed by company names. In contrast, by indexing stocks based on their ranks in capitalization, we gain a different perspective of market dynamics in rank space. Here, we demonstrate the superior performance of statistical arbitrage in rank space over name space, driven by a robust market representation and enhanced mean-reverting properties of residual returns in rank space. Our statistical arbitrage algorithm features an intraday rebalancing mechanism for effective conversion between portfolios in name and rank space. We explore statistical arbitrage with and without neural networks in both name and rank space and show that the portfolios obtained in rank space with neural networks significantly outperform those in name space.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces a novel application of statistical arbitrage by operating in rank space rather than traditional name space, supported by a formal stochastic portfolio theory framework and Ornstein-Uhlenbeck processes. The empirical backtesting across U.S. equities from 2007 to 2022 is rigorous, demonstrating impressive risk-adjusted returns using neural networks combined with an intraday rebalancing scheme.

Research Flowchart

  flowchart TD
    A[Research Goal: Evaluate Statistical Arbitrage in Rank Space vs Name Space] --> B[Data Inputs: Equities Market Capitalization & Intraday Stock Returns]
    B --> C[Methodology: Map Stocks to Rank Space & Extract Residual Returns]
    C --> D[Computational Models: Neural Networks vs Standard StatArb Algorithm]
    D --> E[Execution: Intraday Rebalancing Mechanism between Name and Rank Space]
    E --> F[Findings: Rank Space with NNs achieves Superior Mean Reversion & Portfolio Outperformance]

Cited References (60)

  1. Many-particle physics (2013)
  2. Stochastic portfolio theory: an overview (2009)
  3. Atlas models of equity markets (2005)
  4. Hybrid atlas models (2011)
  5. Strong solutions of stochastic equations with rank-based coefficients (2013)
  6. The cross-section of expected stock returns (1992)
  7. A five-factor asset pricing model (2015)
  8. Factor models, machine learning, and asset pricing (2022)
and 52 more
  1. Instrumented principal component analysis (2020)
  2. Deep learning statistical arbitrage (2021) — link
  3. Statistical arbitrage in the US equities market (2010)
  4. Statistical arbitrage: algorithmic trading insights and techniques (2011)
  5. Risk control of mean-reversion time in statistical arbitrage (2017)
  6. Empirical asset pricing via machine learning (2020)
  7. Going deeper with convolutions (2015)
  8. Attention is all you need (2017)
  9. Bert: Pre-training of deep bidirectional transformers for language understanding (2018) — link
  10. Instance normalization: The missing ingredient for fast stylization (2016) — link
  11. Deep residual learning for image recognition (2016)
  12. Improving neural networks by preventing co-adaptation of feature detectors (2012) — link
  13. Layer normalization (2016) — link
  14. Local times of ranked continuous semimartingales (2008)
  15. Foundations of modern probability (1997)
  16. Stationary distributions of the Atlas model (2018)
  17. Stationary gap distributions for infinite systems of competing Brownian particles (2017)
  18. Equilibrium fluctuation of the Atlas model (2017)
  19. Fluctuations of the Atlas model from inhomogeneous stationary profiles (2023) — link
  20. Existence and uniqueness of semimartingale reflecting Brownian motions in convex polyhedrons (1996)
  21. Multidimensional reflected Brownian motions having exponential stationary distributions (1987)
  22. On the stability of the linear Skorohod problem in an orthant (2002)
  23. Positive recurrence of reflecting Brownian motion in three dimensions (2010)
  24. Dynamic pairs trading using the stochastic control approach (2013)
  25. Algorithmic and high-frequency trading (2015)
  26. Optimal timing to purchase options (2011)
  27. Optimizing a portfolio of mean-reverting assets with transaction costs via a feedforward neural network (2020)
  28. Optimizing the Pairs-Trading Strategy Using Deep Reinforcement Learning with Trading and Stop-Loss Boundaries (2019)
  29. Factors that fit the time series and cross-section of stock returns (2020)
  30. Cleaning large correlation matrices: tools from random matrix theory (2017)
  31. Principal eigenportfolios for US equities (2022)
  32. Understanding systematic risk: A high-frequency approach (2020)
  33. Characteristics are covariances: A unified model of risk and return (2019)
  34. Deep learning in asset pricing (2024)
  35. Deep Reinforcement Learning and Mean-Variance Strategies for Responsible Portfolio Optimization (2024) — link
  36. Finbert: Financial sentiment analysis with pre-trained language models. arXiv 2019 (2019) — link
  37. Deep learning volatility: a deep neural network perspective on pricing and calibration in (rough) volatility models (2021)
  38. Stochastic controls: Hamiltonian systems and HJB equations (2012)
  39. Adam: A method for stochastic optimization (2014)
  40. The Data Stabilizing Transformation from Named to Ranked Equity Returns (2025)
  41. Stochastic portfolio theory: an overview (2009)
  42. Stochastic portfolio theory (2002)
  43. Determining the number of factors from empirical distribution of eigenvalues (2010)
  44. Optimal mean reversion trading: Mathematical analysis and practical applications (2015)
  45. Maximum Likelihood Estimation of a Mean Reverting Process (2023)
  46. Pairs trading (2005)
  47. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations (2019)
  48. Random walks in biology (1993)
  49. Sociophysics: A Physicist’s Modeling of Psycho-Political Phenomena (Understanding Complex Systems) (2013)
  50. Noisy covariance matrices and portfolio optimization (2002)
  51. Noisy covariance matrices and portfolio optimization II (2003)
  52. Financial applications of random matrix theory: a short review (2009) — link

0 of 60 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.