pith. sign in

arxiv: 2511.01471 · v2 · submitted 2025-11-03 · 💱 q-fin.CP · cs.NA· math.NA· q-fin.TR

Trade Execution Flow as the Underlying Source of Market Dynamics

Mikhail Gennadievich Belov , Victor Victorovich Dubov , Vadim Konstantinovich Ivanov , Alexander Yurievich Maslov , Olga Vladimirovna Proshina , Vladislav Gennadievich Malyshkin This is my paper

Pith reviewed 2026-05-18 01:57 UTC · model grok-4.3

classification 💱 q-fin.CP cs.NAmath.NAq-fin.TR
keywords execution flowmarket dynamicsRadon-Nikodym derivativeChristoffel functiontrade executionquantitative financeeigenproblemPCA alternative
0
0 comments X

The pith

Execution flow, the time derivative of traded volume, drives observed market price dynamics according to experiments on real data.

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

The paper sets out to show that market price movements are caused primarily by execution flow rather than other factors. Execution flow is defined as the rate I equals dV over dt, where V is volume, and the authors compute it directly from transaction records by applying the Radon-Nikodym derivative. This calculation also yields natural thresholds for trading decisions and a characteristic time scale extracted from an associated eigenproblem. The same framework introduces a Christoffel-function spectrum that stays unchanged under any non-singular linear remapping of the input variables, providing an alternative to principal-component analysis.

Core claim

The central discovery is that execution flow I = dV/dt, obtained via the Radon-Nikodym derivative applied to raw trade data, constitutes the fundamental driver of market dynamics; validation on actual market records confirms that this flow supplies actionable thresholds and an intrinsic time scale from the corresponding eigenproblem, while the accompanying Christoffel-function spectrum remains invariant under arbitrary non-degenerate linear transformations of the attributes.

What carries the argument

Execution flow I = dV/dt isolated by the Radon-Nikodym derivative on transaction data, together with the Christoffel function spectrum for linear-invariant analysis.

If this is right

  • Trading systems can monitor computed execution-flow thresholds to generate signals without external tuning.
  • Market models should treat execution flow as the causal input variable rather than price or order-book snapshots.
  • Characteristic time scales of market activity can be read directly from the eigenproblem of the flow operator.
  • The Christoffel spectrum supplies a stable substitute for PCA when attributes undergo linear rescaling.

Where Pith is reading between the lines

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

  • If execution flow is causal, then interventions that alter flow (such as order-size caps) should produce measurable changes in subsequent volatility.
  • The same derivative technique could be tested on non-equity markets to check whether flow remains the dominant driver across asset classes.
  • A direct comparison of flow-derived time scales against those obtained from standard autocorrelation studies would clarify whether the eigenproblem adds new information.

Load-bearing premise

The Radon-Nikodym derivative applied to raw market data isolates the true causal execution flow instead of merely correlating with price changes.

What would settle it

A large dataset in which substantial price movements occur with near-zero execution flow, or large flows produce no price response, would disprove the central claim.

Figures

Figures reproduced from arXiv: 2511.01471 by Alexander Yurievich Maslov, Mikhail Gennadievich Belov, Olga Vladimirovna Proshina, Vadim Konstantinovich Ivanov, Victor Victorovich Dubov, Vladislav Gennadievich Malyshkin.

Figure 1
Figure 1. Figure 1: FIG. 1. An example of regular exponential moving average [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. An example of a higher-order orthogonal polynomial [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. A demonstration of execution flow. We present the [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. A demonstration of P&L calculation according to [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. The directional information ( [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. The [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. A presentation of [PITH_FULL_IMAGE:figures/full_fig_p014_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. A presentation of [PITH_FULL_IMAGE:figures/full_fig_p015_8.png] view at source ↗
read the original abstract

In this work, we demonstrate experimentally that the execution flow, $I = dV/dt$, is the fundamental driving force of market dynamics. We develop a numerical framework to calculate execution flow from the data using the Radon-Nikodym derivative. A notable feature of this approach is its ability to automatically determine thresholds that can serve as actionable triggers. The technique also determines the characteristic time scale directly from the corresponding eigenproblem. The methodology has been validated on actual market data to support these findings. Additionally, we introduce a framework based on the Christoffel function spectrum, which is invariant under arbitrary non-degenerate linear transformations of input attributes and offers an alternative to traditional principal component analysis (PCA), which is limited to unitary invariance.

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

3 major / 2 minor

Summary. The paper claims to demonstrate experimentally that the execution flow, I = dV/dt, is the fundamental driving force of market dynamics. It develops a numerical framework using the Radon-Nikodym derivative to calculate this flow from market data, automatically determining thresholds for actionable triggers and characteristic time scales via the corresponding eigenproblem. The methodology is validated on actual market data, and a separate framework based on the Christoffel function spectrum is introduced as an alternative to PCA that is invariant under arbitrary non-degenerate linear transformations of input attributes.

Significance. If substantiated, the identification of execution flow as a causal driver of market dynamics would be a notable contribution to quantitative finance and market microstructure, offering a mechanistic rather than purely statistical account of price formation. The automatic, data-driven extraction of thresholds and time scales could have direct practical utility for trading systems, while the Christoffel spectrum provides a theoretically attractive invariant alternative to PCA. These elements, if supported by rigorous out-of-sample tests and causality checks, would strengthen falsifiable modeling in the field.

major comments (3)
  1. [Abstract] Abstract: The statement that the methodology 'has been validated on actual market data to support these findings' is made without any quantitative results, error metrics, baseline comparisons, or explicit description of how the Radon-Nikodym framework isolates causality rather than correlation. This omission is load-bearing for the central claim that execution flow is the fundamental driving force.
  2. [Abstract] Abstract (framework description): Thresholds for actionable triggers and the characteristic time scale are extracted directly from the same Radon-Nikodym derivative and eigenproblem applied to the validation dataset. This setup risks reducing the driving-force demonstration to an in-sample fit rather than an independent test, as both the model parameters and the validation rely on the identical sequence of trades and quotes.
  3. [Christoffel function spectrum] Christoffel function spectrum section: While the spectrum is stated to be invariant under linear transformations, no explicit demonstration is provided that it mediates or strengthens the claimed causal relation between execution flow I = dV/dt and observed market dynamics; the construction appears decoupled from the main empirical validation.
minor comments (2)
  1. [Abstract] The notation I = dV/dt for execution flow should be accompanied by a precise definition in terms of trade volume and time stamps to avoid ambiguity with standard order-flow measures.
  2. [Introduction] Additional references to existing literature on Radon-Nikodym derivatives in stochastic processes or market-microstructure models would help situate the numerical framework.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their careful reading and valuable comments on our manuscript. We respond to each of the major comments below, indicating where revisions will be made to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The statement that the methodology 'has been validated on actual market data to support these findings' is made without any quantitative results, error metrics, baseline comparisons, or explicit description of how the Radon-Nikodym framework isolates causality rather than correlation. This omission is load-bearing for the central claim that execution flow is the fundamental driving force.

    Authors: The abstract is intended as a concise overview, with detailed quantitative results, error metrics, and baseline comparisons provided in the main body of the paper (specifically in the empirical validation sections). Regarding the isolation of causality, the Radon-Nikodym derivative framework models the execution flow as the source term in the dynamics, and the validation shows superior explanatory power compared to correlation-based approaches. We will revise the abstract to reference these specific results and briefly describe the causal aspect. revision: yes

  2. Referee: [Abstract] Abstract (framework description): Thresholds for actionable triggers and the characteristic time scale are extracted directly from the same Radon-Nikodym derivative and eigenproblem applied to the validation dataset. This setup risks reducing the driving-force demonstration to an in-sample fit rather than an independent test, as both the model parameters and the validation rely on the identical sequence of trades and quotes.

    Authors: While the eigenproblem is solved on the dataset to extract the time scales and thresholds, the demonstration of execution flow as the driving force is validated by applying these to predict and explain the market dynamics in a way that tests the mechanistic link, including checks on different time periods within the data. We acknowledge the potential concern and will add text to clarify that the validation involves testing the predictive implications on the observed price processes, distinguishing it from mere parameter fitting. revision: partial

  3. Referee: [Christoffel function spectrum] Christoffel function spectrum section: While the spectrum is stated to be invariant under linear transformations, no explicit demonstration is provided that it mediates or strengthens the claimed causal relation between execution flow I = dV/dt and observed market dynamics; the construction appears decoupled from the main empirical validation.

    Authors: The Christoffel spectrum is presented as an alternative analytical tool with desirable invariance properties. To address the decoupling, we will revise the section to include an explicit example or analysis showing how the spectrum applied to the execution flow data yields consistent insights into the market dynamics, thereby reinforcing the main claims. revision: yes

Circularity Check

1 steps flagged

Data-derived thresholds and eigenproblem time scales in Radon-Nikodym framework make validation of I=dV/dt as driver dependent on same-data fitting

specific steps
  1. fitted input called prediction [Abstract]
    "A notable feature of this approach is its ability to automatically determine thresholds that can serve as actionable triggers. The technique also determines the characteristic time scale directly from the corresponding eigenproblem. The methodology has been validated on actual market data to support these findings."

    Thresholds and time scale are computed directly from the market data inside the Radon-Nikodym/eigenproblem pipeline; the same pipeline is then used to validate that the extracted I=dV/dt is the causal driver. The 'validation' therefore operates on quantities whose selection was already optimized to the target dataset, making the support for the driving-force claim statistically forced rather than independently derived.

full rationale

The paper's central experimental claim rests on a numerical framework that extracts execution flow via Radon-Nikodym derivative while automatically selecting thresholds and characteristic time scales from the eigenproblem solved on the identical market dataset used for validation. Because both the flow extraction and the supporting parameters are obtained from the same raw trades/quotes without reported out-of-sample hold-out, interventional null, or conditioning on lagged prices, the demonstration that flow is the 'fundamental driving force' reduces to a data-tuned correlation rather than an independent test. The Christoffel-spectrum component is presented as an invariant alternative to PCA but does not mediate or falsify the driving-force relation. This matches the fitted-input-called-prediction pattern at the validation step.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 1 invented entities

Based solely on the abstract, the central claim rests on the domain assumption that execution flow can be isolated as causal via the described numerical tools and that real-market validation confirms this without circular data use; limited information prevents exhaustive listing.

free parameters (2)
  • thresholds for actionable triggers
    Automatically determined by the Radon-Nikodym approach from data
  • characteristic time scale
    Extracted from the eigenproblem solution on market data
axioms (2)
  • domain assumption Execution flow I = dV/dt is the fundamental driving force of market dynamics
    Central experimental claim stated in the abstract
  • domain assumption Radon-Nikodym derivative applied to trading data yields meaningful execution flow and triggers
    Basis for the numerical framework described
invented entities (1)
  • Christoffel function spectrum no independent evidence
    purpose: Invariant alternative to PCA under arbitrary non-degenerate linear transformations
    New framework introduced in the abstract for attribute analysis

pith-pipeline@v0.9.0 · 5690 in / 1660 out tokens · 43594 ms · 2026-05-18T01:57:46.205132+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

29 extracted references · 29 canonical work pages · 4 internal anchors

  1. [1]

    and followed by others [10, 11] among many others – have explored the performance characteristics of a vari- ety of automated trading systems. While our group has previously conducted high-frequency trading (HFT) on NASDAQ, the present study focuses primarily on market microstructure analysis, emphasizing execution flow as the fundamental driving mechanis...

    work page

  2. [2]

    Development of a fast and numerically stable method for moment calculation (Section II)

    work page

  3. [3]

    Application of this method to real exchange data (Section III)

    work page

  4. [4]

    The most important result is the auto- matic determination of the characteristic time scale from the corresponding eigenproblem

    Development of an execution flow estimation methodology (Section IV) and experimental evi- dence linking execution flow singularities to price singularities. The most important result is the auto- matic determination of the characteristic time scale from the corresponding eigenproblem

    work page

  5. [5]

    Derivation of a procedure for converting execution flow fluctuations into probabilistic forecasts of price changes (Sections V and VI)

    work page

  6. [6]

    secondary sampling

    Empirical comparison of the derived directional in- formation with observed market behavior (Section VII). Additionally, we propose a framework based on the Christoffel function spectrum for determining probability contribution components (Appendix C),which is invariant under arbitrary non-degenerate transformations of input attributes. This invariance pr...

    work page 2008

  7. [7]

    Polanyi, Aristotle discovers the economy, Trade and market in the early empires , 64 (1957)

    K. Polanyi, Aristotle discovers the economy, Trade and market in the early empires , 64 (1957)

    work page 1957

  8. [8]

    Walras,Elements of pure economics: Or the theory of social wealth(Routledge, 2013)

    L. Walras,Elements of pure economics: Or the theory of social wealth(Routledge, 2013)

    work page 2013

  9. [9]

    Donier and J.-P

    J. Donier and J.-P. Bouchaud, From Walras’ auctioneer to continuous time double auctions: A general dynamic theory of supply and demand, Journal of Statistical Me- chanics: Theory and Experiment2016, 123406 (2016)

    work page 2016

  10. [10]

    V. G. Malyshkin, Market Dynamics. On A Muse Of Cash Flow And Liquidity Deficit, ArXiv e-prints 10.48550/arXiv.1709.06759 (2017), arXiv:1709.06759 [q- fin.TR]

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1709.06759 2017

  11. [11]

    V. G. Malyshkin, Market Dynamics: On Directional Information Derived From (Time, Execution Price, Shares Traded) Transaction Sequences, arXiv preprint arXiv:1903.11530 10.48550/arXiv.1903.11530 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1903.11530 1903

  12. [12]

    V. G. Malyshkin and R. Bakhramov, Mathematical Foun- dations of Realtime Equity Trading. Liquidity Deficit and Market Dynamics. Automated Trading Machines, arXiv preprint arXiv:1510.05510 10.48550/arXiv.1510.05510 (2015)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1510.05510 2015

  13. [13]

    V. G. Malyshkin, Market Dynamics. On Supply and De- mand Concepts, ArXiv e-prints (2016), http://arxiv. org/abs/1602.04423, arXiv:1602.04423

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  14. [14]

    Bucci, M

    F. Bucci, M. Benzaquen, F. Lillo, and J.-P. Bouchaud, Crossover from linear to square-root market impact, Phys- ical review letters122, 108302 (2019)

    work page 2019

  15. [15]

    Kearns and L

    M. Kearns and L. Ortiz, The Penn-Lehman automated trading project, IEEE Intelligent systems18, 22 (2003)

    work page 2003

  16. [16]

    LeBaron, Agent-based computational finance, Hand- book of computational economics2, 1187 (2006)

    B. LeBaron, Agent-based computational finance, Hand- book of computational economics2, 1187 (2006)

    work page 2006

  17. [17]

    J. B. Chakole, M. S. Kolhe, G. D. Mahapurush, A. Yadav, and M. P. Kurhekar, A Q-learning agent for automated trading in equity stock markets, Expert Systems with Applications163, 113761 (2021)

    work page 2021

  18. [18]

    V. G. Malyshkin, The code for polynomials calcu- lation (2014), http://www.ioffe.ru/LNEPS/malyshkin/ code.htmland an alternative location

    work page 2014

  19. [19]

    V. G. Malyshkin and M. G. Belov, Market Direc- tional Information Derived From (Time, Execution Price, Shares Traded) Sequence of Transactions. On The Im- pact From The Future, arXiv preprint arXiv:2210.04223 10.48550/arXiv.2210.04223 (2022)

    work page doi:10.48550/arxiv.2210.04223 2022

  20. [20]

    Nasdaq OMX,NASDAQ TotalView-ITCH 4.1, Report (Nasdaq OMX, 2014) see sample data files athttps:// emi.nasdaq.com/ITCH/ and newest version specification TotalView-ITCH 5.0

    work page 2014

  21. [21]

    HautschandR

    N. HautschandR. Huang,Limitorderflow,marketimpact and optimal order sizes: Evidence from nasdaq totalview- itch data (2011)

    work page 2011

  22. [22]

    NYSE,Daily TAQ Client Spec v4.2, Report (NYSE,

    work page

  23. [23]

    see sample data files athttps://ftp.nyse.com/ Historical%20Data%20Samples/DAILY%20TAQ/, which provide two days of free data every quarter

    work page

  24. [24]

    Totik, Orthogonal Polynomials, Surveys in Approxi- mation Theory1, 70 (11 Nov

    V. Totik, Orthogonal Polynomials, Surveys in Approxi- mation Theory1, 70 (11 Nov. 2005)

    work page 2005

  25. [25]

    V. G. Malyshkin, On The Radon-Nikodym Spectral Approach With Optimal Clustering, arXiv preprint arXiv:1906.00460 10.48550/arXiv.1906.00460 (2019)

    work page doi:10.48550/arxiv.1906.00460 1906

  26. [26]

    Lasserre and E

    J.-B. Lasserre and E. Pauwels, The empirical Christoffel function with applications in data analysis, Advances in Computational Mathematics , 1 (2019)

    work page 2019

  27. [27]

    J. B. Lasserre, A disintegration of the Christoffel function, Comptes Rendus. Mathématique360, 1071 (2022)

    work page 2022

  28. [28]

    Lasserre,Moments, positive polynomials and their applications, Vol

    J.-B. Lasserre,Moments, positive polynomials and their applications, Vol. 1 (World Scientific, 2009)

    work page 2009

  29. [29]

    V. G. Malyshkin,On Lebesgue Integral Quadrature,arXiv preprint arXiv:1807.06007 10.48550/arXiv.1807.06007 (2018)

    work page doi:10.48550/arxiv.1807.06007 2018