pith. sign in

arxiv: 2506.07993 · v3 · submitted 2025-06-09 · 💱 q-fin.MF · math.PR

Stochastic portfolio theory with price impact

David Itkin This is my paper

Pith reviewed 2026-05-19 11:24 UTC · model grok-4.3

classification 💱 q-fin.MF math.PR
keywords stochastic portfolio theoryprice impactfunctional generationrelative arbitragemarket portfolioimpact decaysemimartingale prices
0
0 comments X

The pith

Stochastic portfolio theory extends to nonlinear price impact by deriving the master formula for additive functional generation of trading strategies.

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

The paper builds a framework for stochastic portfolio theory that includes nonlinear price impact and impact decay in high-dimensional markets. It shows that the master formula for generating trading strategies from portfolio-generating functions carries over from the frictionless case, yielding explicit expressions for relative wealth, positive price conditions, and an SDE for the observed price process together with holdings and impact states. The work also gives relative arbitrage conditions and applies the results to backtests of quadratic and entropy portfolios on equity data, where impact reduces performance. A reader would care because real-world trading moves prices, so this makes functional generation usable beyond idealized models.

Core claim

The central claim is that the celebrated master formula for additive functional generation of trading strategies extends directly to a general high-dimensional market model with nonlinear price impact and decaying impact kernels, while the observed price remains a semimartingale and the usual relative-wealth and arbitrage formulas continue to hold after suitable adjustment for impact state processes.

What carries the argument

The master formula for additive functional generation of trading strategies, extended to track price-impact state processes alongside observed prices and holdings.

If this is right

  • Relative wealth of a functionally generated portfolio versus the market portfolio is given by an explicit formula that includes impact effects.
  • Conditions on the generating function guarantee that observed market prices remain positive.
  • An SDE describes the joint evolution of observed prices, investor holdings, and price-impact state variables.
  • Relative arbitrage opportunities exist under the same functional-generation conditions as in the frictionless case, now adjusted for impact.
  • Backtests of quadratic and entropy generating functions on US equity data show measurable performance erosion once price impact is included.

Where Pith is reading between the lines

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

  • The framework suggests that high-turnover functional portfolios will require explicit impact adjustments to preserve claimed outperformance in live trading.
  • Similar impact-adjusted master formulas could be derived for other SPT objects such as functionally generated benchmarks or diversity measures.
  • Empirical work could test whether the derived SDE for observed prices matches transaction-level data when impact kernels are calibrated separately.

Load-bearing premise

Price impact is nonlinear and decays so that the observed price process stays a semimartingale without extra regularity conditions on the impact kernel.

What would settle it

A direct numerical check or market simulation in which the derived SDE for observed prices, holdings, and impact states fails to reproduce the price path produced by the known impact function would falsify the master-formula extension.

Figures

Figures reproduced from arXiv: 2506.07993 by David Itkin.

Figure 1
Figure 1. Figure 1: Results of the numerical experiments across [PITH_FULL_IMAGE:figures/full_fig_p023_1.png] view at source ↗
read the original abstract

We develop a framework for stochastic portfolio theory (SPT), which incorporates modern nonlinear price impact and impact decay models. Our main result is the derivation of the celebrated master formula for additive functional generation of trading strategies in a general high-dimensional market model with price impact. We also derive formulas for an investor's relative wealth with respect to the market portfolio, conditions that guarantee positive observed market prices and a stochastic differential equation governing the dynamics of the observed price, the investor's holdings and the price impact state processes. As an application of these results, we develop conditions for relative arbitrage in the price impact setting analogous to previously obtained results for the frictionless setting. We then apply our framework to backtest the quadratic and entropy generating functions on historical US equity data, illustrating how price impact can negatively affect portfolio performance.

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 / 2 minor

Summary. The paper extends stochastic portfolio theory to markets with nonlinear price impact and decay. Its central claim is that the master formula for additive functional generation of trading strategies carries over to this setting, along with derivations of relative wealth processes, conditions ensuring positive observed prices, an SDE for the joint dynamics of prices, holdings, and impact states, and relative arbitrage conditions. The framework is applied to backtests of quadratic and entropy generating functions on historical US equity data, showing performance degradation due to impact.

Significance. If the extension of the master formula is valid, the work meaningfully advances SPT by incorporating realistic frictions, enabling more applicable relative arbitrage results and performance analysis. The combination of theoretical derivations with an empirical backtest on real data is a strength, as is the focus on high-dimensional settings. The results could inform practical portfolio construction under impact, provided the modeling assumptions hold.

major comments (2)
  1. [§3] §3 (derivation of the master formula): The extension of the frictionless master formula relies on substituting the observed price process while assuming that the nonlinear impact and decay kernel preserve the semimartingale property and allow the functional Itô formula to apply without additional regularity (e.g., Lipschitz continuity or integrable decay rates on the kernel). This is invoked as a modeling choice rather than derived, leaving open the possibility that high-dimensional cross terms produce unaccounted quadratic covariation or finite-variation components not canceled by integration-by-parts.
  2. [§4] §4 (SDE for observed prices and impact states): The coupled SDE for the observed price process, investor holdings, and price impact state is presented without explicit conditions guaranteeing that the impact-driven terms remain semimartingales when interacting with the underlying market semimartingales; this assumption is load-bearing for the subsequent relative wealth and arbitrage formulas.
minor comments (2)
  1. [backtest section] The backtest section would benefit from explicit robustness checks or error bars on the choice of impact parameters (decay rate and nonlinearity), as performance degradation is reported but sensitivity is not quantified.
  2. [§2] Notation for the impact kernel and state process could be clarified with a dedicated table or diagram showing the coupling to cumulative trades.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and valuable comments on our paper. We address the major comments point by point below, providing clarifications and indicating planned revisions to strengthen the manuscript.

read point-by-point responses

  1. Referee: [§3] §3 (derivation of the master formula): The extension of the frictionless master formula relies on substituting the observed price process while assuming that the nonlinear impact and decay kernel preserve the semimartingale property and allow the functional Itô formula to apply without additional regularity (e.g., Lipschitz continuity or integrable decay rates on the kernel). This is invoked as a modeling choice rather than derived, leaving open the possibility that high-dimensional cross terms produce unaccounted quadratic covariation or finite-variation components not canceled by integration-by-parts.

    Authors: We appreciate the referee's observation regarding the assumptions underlying the master formula derivation. In the paper, the observed price is constructed by adding the impact term to the unaffected semimartingale price process, and we model the impact using a nonlinear function applied to the trading process convolved with a decay kernel. The semimartingale property is preserved by assumption, consistent with standard approaches in the price impact literature (e.g., models with transient impact). Regarding potential unaccounted quadratic covariation in high dimensions, the functional Itô formula is applied to the relative wealth functional, and cross terms are handled through the quadratic variation of the observed price process, which incorporates the impact effects. However, to make this more rigorous, we will revise the manuscript to explicitly state the required regularity conditions on the impact function and decay kernel, such as Lipschitz continuity and integrability, ensuring the applicability of the Itô formula without additional finite-variation terms. revision: partial

  2. Referee: [§4] §4 (SDE for observed prices and impact states): The coupled SDE for the observed price process, investor holdings, and price impact state is presented without explicit conditions guaranteeing that the impact-driven terms remain semimartingales when interacting with the underlying market semimartingales; this assumption is load-bearing for the subsequent relative wealth and arbitrage formulas.

    Authors: We thank the referee for pointing this out. The coupled SDE is derived by differentiating the definitions of the observed price and impact state, assuming that the impact function is differentiable and that the resulting processes inherit the semimartingale property from the underlying market and trading processes. This assumption is indeed crucial for the validity of the relative wealth and arbitrage results that follow. To address the concern, we will add a proposition or remark in the revised version that provides sufficient conditions (e.g., bounded variation of the decay kernel and Lipschitz continuity of the nonlinear impact map) under which the impact-driven terms remain semimartingales. This will make the load-bearing assumption explicit and verifiable. revision: yes

Circularity Check

0 steps flagged

Master formula derived from semimartingale dynamics with impact; no reduction to inputs by construction or self-citation.

full rationale

The paper derives the master formula by substituting the observed price SDE (coupled to impact state and trades) into the standard functional Itô calculus for additive generation, under the modeling assumption that nonlinear impact and decay preserve the semimartingale property. This is an extension step with independent content rather than a tautology; the backtests apply standard generating functions (quadratic, entropy) to data without fitting parameters to force the claimed performance. No self-citation load-bearing on uniqueness theorems, no fitted inputs renamed as predictions, and no ansatz smuggled via prior work. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The framework rests on standard stochastic calculus assumptions plus domain-specific modeling choices for impact; no new particles or forces are invented, but several modeling parameters for impact are introduced.

free parameters (1)
  • impact decay rate and nonlinearity parameters
    Chosen to model temporary and permanent impact; values are not derived from first principles and affect the observed price SDE and backtest outcomes.
axioms (1)
  • domain assumption The price impact process is adapted and the resulting observed price remains a semimartingale.
    Invoked to ensure the functional generation approach and Ito calculus extensions remain valid.

pith-pipeline@v0.9.0 · 5649 in / 1425 out tokens · 36707 ms · 2026-05-19T11:24:23.755441+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

42 extracted references · 42 canonical work pages

  1. [1]

    Abi Jaber and E

    E. Abi Jaber and E. Neuman. Optimal liquidation with signals: the general propagator case. Forthcoming in Mathe- matical Finance, 2025

    work page 2025

  2. [2]

    Fredholm approach to nonlinear propagator models

    Eduardo Abi Jaber, Alessandro Bondi, Nathan De Carvalho, Eyal Neuman, and Sturmius Tuschmann. Fredholm approach to nonlinear propagator models. Available at SSRN 5167824 , 2025

    work page 2025

  3. [3]

    C` adl` ag semimartingale strategies for optimal trade execution in stochastic order book models

    Julia Ackermann, Thomas Kruse, and Mikhail Urusov. C` adl` ag semimartingale strategies for optimal trade execution in stochastic order book models. Finance and Stochastics , 25:757–810, 2021

    work page 2021

  4. [4]

    Alfonsi, A

    A. Alfonsi, A. Fruth, and A. Schied. Optimal execution strategies in limit order books with general shape functions. Quantitative Finance , 10(2):143–157, 2010

    work page 2010

  5. [5]

    Almgren and N

    R. Almgren and N. Chriss. Optimal execution of portfolio transactions. Journal of Risk , 3:5–39, 2000

    work page 2000

  6. [6]

    Almgren, C

    R. Almgren, C. Thum, E. Hauptmann, and H. Li. Direct estimation of equity market impact. Risk, 18(7):57–62, 2005

    work page 2005

  7. [7]

    Banner, Robert Fernholz, and Ioannis Karatzas

    Adrian D. Banner, Robert Fernholz, and Ioannis Karatzas. Atlas models of equity markets. The Annals of Applied Probability, 15(4):2296 – 2330, 2005

    work page 2005

  8. [8]

    Bershova and D

    N. Bershova and D. Rakhlin. The non-linear market impact of large trades: Evidence from buy-side order flow. Quantitative Finance , 13(11):1759–1778, 2013. 38

    work page 2013

  9. [9]

    Bouchaud, J

    J.-P. Bouchaud, J. Bonart, J. Donier, and M. Gould. Trades, Quotes and Prices . Cambridge University Press, Cambridge, UK, 2018

    work page 2018

  10. [10]

    Bouchaud, Y

    J.-P. Bouchaud, Y. Gefen, M. Potters, and M. Wyart. Fluctuations and response in financial markets: the subtle nature of ‘random’ price changes. Quantitative Finance , 4(2):176, 2003

    work page 2003

  11. [11]

    Brokmann, E

    X. Brokmann, E. S´ eri´ e, J. Kockelkoren, and J.-P. Bouchaud. Slow decay of impact in equity markets. Market Mi- crostructure and Liquidity , 1(2):1550007, 2015

    work page 2015

  12. [12]

    Efficient trading with price impact

    Xavier Brokmann, Lukas Gonon, Guangyi He, David Itkin, and Johannes Muhle-Karbe. Efficient trading with price impact. Available at SSRN , 2024

    work page 2024

  13. [13]

    Functional portfolio optimization in stochastic portfolio theory

    Steven Campbell and Ting-Kam Leonard Wong. Functional portfolio optimization in stochastic portfolio theory. SIAM Journal on Financial Mathematics , 13(2):576–618, 2022

    work page 2022

  14. [14]

    Macroscopic properties of equity markets: stylized facts and portfolio performance

    Steven Campbell, Qien Song, and Ting-Kam Leonard Wong. Macroscopic properties of equity markets: stylized facts and portfolio performance. arXiv preprint arXiv:2409.10859 , 2024

    work page arXiv 2024

  15. [15]

    Cartea, S

    ´A. Cartea, S. Jaimungal, and J. Penalva. Algorithmic and High-Frequency Trading . Cambridge University Press, Cambridge, UK, 2015

    work page 2015

  16. [16]

    Polynomial processes in stochastic portfolio theory

    Christa Cuchiero. Polynomial processes in stochastic portfolio theory. Stochastic processes and their applications , 129(5):1829–1872, 2019

    work page 2019

  17. [17]

    Signature methods in stochastic portfolio theory.arXiv preprint arXiv:2310.02322 , 2023

    Christa Cuchiero and Janka M¨ oller. Signature methods in stochastic portfolio theory.arXiv preprint arXiv:2310.02322 , 2023

    work page arXiv 2023

  18. [18]

    Robert Fernholz

    E. Robert Fernholz. Stochastic portfolio theory , volume 48 of Applications of Mathematics (New York) . Springer- Verlag, New York, 2002. Stochastic Modelling and Applied Probability

    work page 2002

  19. [19]

    Robert Fernholz, Ioannis Karatzas, and Johannes Ruf

    E. Robert Fernholz, Ioannis Karatzas, and Johannes Ruf. Volatility and arbitrage. Ann. Appl. Probab. , 28(1):378–417, 2018

    work page 2018

  20. [20]

    On the diversity of equity markets

    Robert Fernholz. On the diversity of equity markets. Journal of Mathematical Economics , 31(3):393–417, 1999

    work page 1999

  21. [21]

    Portfolio generating functions

    Robert Fernholz. Portfolio generating functions. In Quantitative Analysis in Financial Markets: Collected Papers of the New York University Mathematical Finance Seminar , pages 344–367. World Scientific, 1999

    work page 1999

  22. [22]

    Relative arbitrage in volatility-stabilized markets

    Robert Fernholz and Ioannis Karatzas. Relative arbitrage in volatility-stabilized markets. Annals of Finance , 1:149– 177, 2005

    work page 2005

  23. [23]

    Diversity and relative arbitrage in equity markets

    Robert Fernholz, Ioannis Karatzas, and Constantinos Kardaras. Diversity and relative arbitrage in equity markets. Finance and Stochastics , 9:1–27, 2005

    work page 2005

  24. [24]

    Stochastic portfolio theory and stock market equilibrium

    Robert Fernholz and Brian Shay. Stochastic portfolio theory and stock market equilibrium. The Journal of Finance , 37(2):615–624, 1982

    work page 1982

  25. [25]

    Frazzini, R

    A. Frazzini, R. Israel, and T. J. Moskowitz. Trading costs. Preprint, 2018

    work page 2018

  26. [26]

    Gˆ arleanu and L

    N. Gˆ arleanu and L. H. Pedersen. Dynamic portfolio choice with frictions. Journal of Economic Theory , 165:487–516, 2016

    work page 2016

  27. [27]

    Hedging, arbitrage and optimality with superlinear frictions

    Paolo Guasoni and Mikl´ os R´ asonyi. Hedging, arbitrage and optimality with superlinear frictions. The Annals of Applied Probability, 25(4):2066 – 2095, 2015

    work page 2066

  28. [28]

    N. Hey, I. Mastromatteo, J. Muhle-Karbe, and K. Webster. Trading with concave price impact – theory and evidence. Preprint, 2023

    work page 2023

  29. [29]

    Concave cross impact

    Natascha Hey, Iacopo Mastromatteo, and Johannes Muhle-Karbe. Concave cross impact. Available at SSRN , 2024

    work page 2024

  30. [30]

    Horn and Charles R

    Roger A. Horn and Charles R. Johnson. Matrix analysis . Cambridge University Press, Cambridge, second edition, 2013

    work page 2013

  31. [31]

    Hybrid Atlas models

    Tomoyuki Ichiba, Vassilios Papathanakos, Adrian Banner, Ioannis Karatzas, and Robert Fernholz. Hybrid Atlas models. The Annals of Applied Probability , 21(2):609 – 644, 2011

    work page 2011

  32. [32]

    Open markets and hybrid jacobi processes

    David Itkin and Martin Larsson. Open markets and hybrid jacobi processes. The Annals of Applied Probability , 34(3):2940–2985, 2024

    work page 2024

  33. [33]

    Trading strategies generated by lyapunov functions

    Ioannis Karatzas and Johannes Ruf. Trading strategies generated by lyapunov functions. Finance and Stochastics , 21(3):753–787, 2017

    work page 2017

  34. [34]

    Relative arbitrage: Sharp time horizons and motion by curvature

    Martin Larsson and Johannes Ruf. Relative arbitrage: Sharp time horizons and motion by curvature. Mathematical Finance, 31(3):885–906, 2021

    work page 2021

  35. [35]

    Statistical arbitrage in rank space

    Y-F Li and George Papanicolaou. Statistical arbitrage in rank space. arXiv preprint arXiv:2410.06568 , 2024

    work page arXiv 2024

  36. [36]

    Almada Monter, Mykhaylo Shkolnikov, and Jiacheng Zhang

    Sergio A. Almada Monter, Mykhaylo Shkolnikov, and Jiacheng Zhang. Dynamics of observables in rank-based models and performance of functionally generated portfolios. The Annals of Applied Probability , 29(5):2849 – 2883, 2019

    work page 2019

  37. [37]

    Obizhaeva and J

    A. Obizhaeva and J. Wang. Optimal trading strategy and supply/demand dynamics. Journal of Financial Markets , 16(1):1–32, 2013

    work page 2013

  38. [38]

    The geometry of relative arbitrage

    Soumik Pal and Ting-Kam Leonard Wong. The geometry of relative arbitrage. Mathematics and Financial Economics , 10:263–293, 2016

    work page 2016

  39. [39]

    Continuous martingales and Brownian motion , volume 293 of Grundlehren der mathe- matischen Wissenschaften [Fundamental Principles of Mathematical Sciences]

    Daniel Revuz and Marc Yor. Continuous martingales and Brownian motion , volume 293 of Grundlehren der mathe- matischen Wissenschaften [Fundamental Principles of Mathematical Sciences] . Springer-Verlag, Berlin, third edition, 1999

    work page 1999

  40. [40]

    The impact of proportional transaction costs on systematically generated portfolios

    Johannes Ruf and Kangjianan Xie. The impact of proportional transaction costs on systematically generated portfolios. SIAM Journal on Financial Mathematics , 11(3):881–896, 2020

    work page 2020

  41. [41]

    K. Webster. Handbook of Price Impact Modeling . CRC Press, Boca Raton, FL, 2023

    work page 2023

  42. [42]

    Zarinelli, M

    E. Zarinelli, M. Treccani, J. D. Farmer, and F. Lillo. Beyond the square root: Evidence for logarithmic dependence of market impact on size and participation rate. Market Microstructure and Liquidity , 1(2):1550004, 2015. 39

    work page 2015