pith. machine review for the scientific record. sign in

arxiv: 2604.18046 · v1 · submitted 2026-04-20 · 💻 cs.CE · cs.MA

Recognition: unknown

EvoMarket: A High-Fidelity and Scalable Financial Market Simulator

Muyao Zhong , Zhenhua Yang , Yuxiang Liu , Ke Tang , Peng Yang
Authors on Pith no claims yet

Pith reviewed 2026-05-10 03:41 UTC · model grok-4.3

classification 💻 cs.CE cs.MA
keywords market simulationlimit order bookself-calibrationmulti-agent simulationdiscrete-event simulationfinancial marketsmarket microstructureChina A-shares
0
0 comments X

The pith

EvoMarket achieves close replay of historical market data over multiple trading days by using an Oracle to add corrective orders when the simulation drifts from real microstructure.

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

EvoMarket is a discrete-event multi-agent simulator built for financial markets that handles multi-asset and cross-day experiments. It incorporates standard market rules such as opening auctions, price limits, and settlement procedures to keep the setting realistic. The central advance is a self-calibration process guided by an Oracle that detects when the simulated limit order book diverges from historical records and inserts synthetic orders to correct the drift at designated checkpoints. This approach removes the need for costly pre-run tuning and produces alignment with actual China A-share order flow and order books across five trading days. The result matters because it supports scalable tests of market interventions, stress scenarios, and policy changes in environments that previous simulators could not sustain at full breadth.

Core claim

EvoMarket couples a high-throughput execution core with optimized limit order book data structures, hierarchical scheduling under delays, and asynchronous per-asset matching together with explicit institutional mechanisms including market calendars, opening call auctions, price limits, and T+1 settlement. It introduces an Oracle-guided in-run self-calibration mechanism that interprets microstructure discrepancies as missing order flow and synthesizes corrective orders at recording checkpoints, yielding close replay alignment over five trading days on China A-share data, fidelity improvements across depth levels, broad agent order coverage, and scalable performance as order rates and market宽度

What carries the argument

Oracle-guided in-run self-calibration mechanism that treats differences between simulated and historical limit order book microstructure as missing order flow and generates synthetic corrective orders at fixed recording checkpoints.

If this is right

  • Enables intervention-oriented experiments across multiple assets and multiple trading days in one system.
  • Delivers measurable fidelity gains from budgeted in-run calibration at varying order-book depth levels.
  • Maintains broad coverage of possible agent order placements during simulation.
  • Preserves performance scalability when input order rates and overall market breadth grow.
  • Produces interpretable event-time responses and cross-asset dependence patterns in event-study style evaluations.

Where Pith is reading between the lines

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

  • The same calibration logic could support counterfactual policy tests by letting an experimenter alter rules mid-run and observe resulting order-flow changes.
  • If the corrective orders remain unbiased, the method may extend to other exchanges or asset classes for comparative stress testing.
  • High throughput suggests possible integration with streaming market data for near-real-time scenario generation.
  • Event-study outputs could help quantify how external shocks transmit through linked assets in multi-market settings.

Load-bearing premise

The Oracle-guided in-run self-calibration mechanism can interpret microstructure discrepancies as missing order flow and synthesize corrective orders at recording checkpoints without introducing systematic biases or artifacts into the simulated market dynamics.

What would settle it

Run the simulator on a held-out multi-day China A-share dataset and measure whether price paths, trade volumes, and order-book depth profiles stay aligned with the real records at multiple time points after each calibration checkpoint, with divergence remaining below a small threshold.

Figures

Figures reproduced from arXiv: 2604.18046 by Ke Tang, Muyao Zhong, Peng Yang, Yuxiang Liu, Zhenhua Yang.

Figure 1
Figure 1. Figure 1: EvoMarket architecture overview. Historical order flow/LOB data, market-calendar mechanisms, and experiment policies drive a discrete-event execution core that routes messages and schedules events, and coordinates multi-asset exchanges and microstructure mechanisms. The Agent Panel hosts heterogeneous financial agent types and provides an interface to extend custom agents. Oracle-guided self-calibration al… view at source ↗
Figure 2
Figure 2. Figure 2: Mid-price alignment under historical order replay for Jan 2–3 (1-minute sampling; pre-open, lunch-break, and overnight intervals omitted). Light blue and light green background bands indicate morning and afternoon sessions; blue vertical lines indicate session boundaries (lunch and day transitions). The inset zooms into the close-to-open transition. Across these two trading days, EvoMarket closely tracks t… view at source ↗
Figure 3
Figure 3. Figure 3: Agent order-space coverage for limit orders (density overlays by agent type). The horizontal axis is the tick offset from mid-price Δtick = (𝑝 − 𝑚)∕Δ𝑝 (symlog); the vertical axis is order size 𝑞 in lots (log) [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Cross-asset correlation heatmaps computed from 1-min mid-price log returns (ABIDES, 20 independent runs treated as 20 assets; EvoMarket, single linked 20-asset run) [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Event-study style intervention evaluation (mid-price only). A step-jump intervention is injected at a fixed time to mimic an aggressive buy/sell sweep. The baseline curve is deterministic under a fixed seed; each intervention direction is repeated 10 times and shown as a mean trajectory after the event. 5.3. Simulation Studies on Microstructure fidelity 5.3.1. In-run self-calibration under fixed compute bu… view at source ↗
Figure 6
Figure 6. Figure 6: Calibration efficiency on one hour of 3-second snapshots. Best-so-far price MSE versus wall-clock time is shown in log-log scale. EvoMarket performs in-run self-calibration and ends after one run, while ABIDES+PSO and MAXE+PSO run an external optimizer for one hour (3,600 seconds), matching the calibration window length [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: One-hour calibration result on 3-second snapshots. Four independent 60-minute cases are visualized at 30-second sampling for readability. Mid-price is measured in RMB. EvoMarket remains close to truth, while ABIDES+PSO and MAXE+PSO show larger residual deviations under the same one-hour wall-clock limit [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Calibrated vs. real LOB at levels 1–5 (bid/ask) over five days (10-minute sampling; pre-open, lunch break, and overnight intervals omitted). Prices (RMB) are shown as lines and Δ𝑉 (shares; calibrated − real) is shown as bars (clipped to ±1𝑒5). Each panel annotates price MSE for the corresponding level. Date ticks indicate day-of-month in January 2019. 5.4. Scalable efficiency and ablations We next validate… view at source ↗
Figure 9
Figure 9. Figure 9: Single-core throughput stress test. Wall-clock time to process a given simulated order rate (orders per second) is reported under increasing load (log-scale x-axis). Lower is better. Values are mean execution times (ms). 1 2 4 8 16 32 64 128 #Stocks 10 0 10 1 Wall-clock (s) workers=1 workers=4 workers=16 workers=64 workers=128 Peak RSS 150 200 250 300 Peak RSS (MB) [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Market-breadth scaling under fixed per-stock order density. We increase the number of stocks while keeping the per-stock order injection rate fixed at 200 orders per second for a 10-second simulated window, and report wall-clock time (left axis, log scale) and peak RSS (right axis). Curves show worker settings 1/4/16/64/128 [PITH_FULL_IMAGE:figures/full_fig_p017_10.png] view at source ↗
read the original abstract

High-fidelity, scalable market simulation is a key instrument for mechanism evaluation, stress testing, and counterfactual policy analysis. Yet existing simulators rarely achieve \emph{mechanism fidelity} beyond single-asset intraday settings, \emph{microstructure fidelity} against historical limit order books (LOB), and \emph{computational tractability} at market scale in a single system. This paper presents \textit{EvoMarket}, a discrete-event, multi-agent financial market simulator designed for intervention-oriented experiments in multi-asset and cross-day environments. EvoMarket couples a high-throughput execution core (optimized LOB data structures, hierarchical scheduling under propagation delays, and asynchronous per-asset matching) with explicit institutional mechanisms (market calendars, opening call auctions, price limits, and T+1 settlement). To avoid expensive black-box calibration, EvoMarket introduces an Oracle-guided in-run self-calibration mechanism that interprets microstructure discrepancy as missing order flow and synthesizes corrective orders at recording checkpoints. Experiments on China A-share order-flow and LOB data show close replay alignment over five trading days, fidelity gains from budgeted in-run calibration across depth levels, broad agent order-space coverage, and scalable performance under increasing input order rates and market breadth. We further demonstrate cross-asset linkage and event-study style intervention evaluation that produces structured dependence and interpretable event-time responses.

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 / 1 minor

Summary. The manuscript presents EvoMarket, a discrete-event multi-agent financial market simulator for multi-asset and cross-day environments. It couples an optimized LOB execution core and explicit institutional mechanisms (calendars, call auctions, price limits, T+1) with an Oracle-guided in-run self-calibration that treats microstructure discrepancies as missing order flow and synthesizes corrective orders at checkpoints. Experiments on China A-share order-flow and LOB data are reported to demonstrate close replay alignment over five trading days, fidelity gains from budgeted calibration, broad agent coverage, scalability with input rates and breadth, plus cross-asset linkage and event-study intervention evaluation.

Significance. If the self-calibration can be shown not to introduce systematic biases into matching, depth evolution, or agent behavior, the simulator would offer a useful platform for mechanism evaluation and counterfactual policy analysis at market scale. The explicit handling of multi-asset and institutional rules addresses a recognized gap in existing simulators; however, the reliance on oracle-driven corrections risks reducing the system to a hybrid replay tool rather than a fully generative model.

major comments (3)
  1. [Abstract] Abstract: The central claim of 'close replay alignment' and 'fidelity gains from budgeted in-run calibration' is presented without any quantitative error metrics (e.g., RMSE on depth profiles, Kolmogorov-Smirnov statistics on order sizes/timings, or ablation results comparing calibrated vs. uncalibrated runs). This absence prevents assessment of whether the alignment is statistically meaningful or merely visual.
  2. [Abstract] Abstract (Oracle-guided in-run self-calibration): The mechanism interprets all LOB discrepancies as missing order flow and synthesizes corrective orders, yet no rule is supplied for choosing correction timing, size, type, or price (e.g., whether limits or T+1 constraints are enforced, or how parameters are sampled to preserve statistical indistinguishability from real flow). Because this step is load-bearing for the fidelity claim, the lack of specification leaves open the possibility that corrections alter subsequent dynamics or mask model deficiencies.
  3. [Abstract] Abstract: The paper advertises utility for 'counterfactual policy analysis' and 'intervention evaluation,' but the oracle calibration depends on historical LOB checkpoints; it is unclear how the system would generate independent trajectories for true counterfactuals without the oracle, undermining the advertised use case.
minor comments (1)
  1. [Abstract] The abstract states 'broad agent order-space coverage' without defining the coverage metric or the agent types employed.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major point below, providing clarifications from the full text and proposing targeted revisions to improve clarity and completeness without altering the core contributions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim of 'close replay alignment' and 'fidelity gains from budgeted in-run calibration' is presented without any quantitative error metrics (e.g., RMSE on depth profiles, Kolmogorov-Smirnov statistics on order sizes/timings, or ablation results comparing calibrated vs. uncalibrated runs). This absence prevents assessment of whether the alignment is statistically meaningful or merely visual.

    Authors: The experimental results section of the manuscript reports quantitative metrics supporting these claims, including RMSE values on depth profiles across multiple levels and Kolmogorov-Smirnov statistics comparing order size and timing distributions between simulated and real data, with explicit ablations showing fidelity gains from calibration. We agree the abstract would benefit from including key quantitative highlights and will revise it accordingly to reference these metrics and their statistical significance. revision: yes

  2. Referee: [Abstract] Abstract (Oracle-guided in-run self-calibration): The mechanism interprets all LOB discrepancies as missing order flow and synthesizes corrective orders, yet no rule is supplied for choosing correction timing, size, type, or price (e.g., whether limits or T+1 constraints are enforced, or how parameters are sampled to preserve statistical indistinguishability from real flow). Because this step is load-bearing for the fidelity claim, the lack of specification leaves open the possibility that corrections alter subsequent dynamics or mask model deficiencies.

    Authors: The methods section details the calibration rules: corrections occur at fixed recording checkpoints, sizes are computed directly from the observed discrepancy volume, order types are selected to match empirical frequencies in the real flow (with limits enforced and T+1 settlement respected), and prices are drawn from the current LOB state or sampled from historical conditional distributions to preserve statistical properties. We will add a concise summary of these rules to the abstract to address the concern. revision: yes

  3. Referee: [Abstract] Abstract: The paper advertises utility for 'counterfactual policy analysis' and 'intervention evaluation,' but the oracle calibration depends on historical LOB checkpoints; it is unclear how the system would generate independent trajectories for true counterfactuals without the oracle, undermining the advertised use case.

    Authors: The simulator architecture supports an independent generative mode in which the oracle and self-calibration are disabled, allowing agents and mechanisms to produce trajectories based solely on their internal models and random seeds. This mode is used for the intervention evaluation experiments described in the results. We will revise the abstract and discussion to explicitly distinguish replay (oracle-enabled) and generative (oracle-disabled) modes to clarify applicability to counterfactual analysis. revision: yes

Circularity Check

0 steps flagged

No circularity: simulator description and empirical alignment claims contain no derivations, equations, or self-referential reductions.

full rationale

The paper introduces a discrete-event multi-agent simulator with an Oracle-guided in-run self-calibration step that synthesizes corrective orders from observed LOB discrepancies. No equations, first-principles derivations, or parameter-fitting procedures are described that would reduce the reported replay alignment to the calibration inputs by construction. The calibration is presented as an external correction mechanism rather than a tautological definition of the target fidelity metric. No self-citations, uniqueness theorems, or ansatzes imported from prior author work appear in the provided text to support core claims. Experimental results are framed as empirical outcomes against external China A-share data, satisfying the self-contained benchmark criterion.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

The central claim rests on standard discrete-event simulation principles plus the novel self-calibration component; no free parameters or axioms are explicitly stated in the abstract.

invented entities (1)
  • Oracle-guided in-run self-calibration no independent evidence
    purpose: Interpret microstructure discrepancies as missing order flow and synthesize corrective orders at checkpoints
    Introduced as the key mechanism to achieve fidelity without expensive black-box calibration.

pith-pipeline@v0.9.0 · 5546 in / 1166 out tokens · 31389 ms · 2026-05-10T03:41:19.540654+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

61 extracted references · 1 canonical work pages

  1. [1]

    Zheng, J

    X. Zheng, J. Li, M. Lu, F.-Y. Wang, New paradigm for economic and financial research with generative ai: Impact and perspective, IEEE Transactions on Computational Social Systems 11 (3) (2024) 3457–3467

    2024

  2. [2]

    Hussain, T

    O. Hussain, T. Dillon, F. K. Hussain, E. Chang, Probabilistic assessment of financial risk in e-business associations, Simulation Modelling Practice and Theory 19 (2) (2011) 704–717

    2011

  3. [3]

    C. Daah, A. Qureshi, I. Awan, S. Konur, Simulation-based evaluation of advanced threat detection and response in financial industry networks using zero trust and blockchain technology, Simulation Modelling Practice and Theory 138 (2025) 103027

    2025

  4. [4]

    Hasbrouck, Empirical market microstructure: The institutions, economics, and econometrics of securities trading, Oxford University Press, 2007

    J. Hasbrouck, Empirical market microstructure: The institutions, economics, and econometrics of securities trading, Oxford University Press, 2007

    2007

  5. [5]

    De Natale, G

    L. De Natale, G. Fargetta, L. R. Scrimali, S. Battiato, Multi-agent reinforcement learning and variational inequality models for international trade networks under crisis, Simulation Modelling Practice and Theory 146 (2026) 103219

    2026

  6. [6]

    Allen, D

    F. Allen, D. Gale, Financial contagion, Journal of Political Economy 108 (1) (2000) 1–33

    2000

  7. [7]

    M. K. Brunnermeier, L. H. Pedersen, Market liquidity and funding liquidity, The Review of Financial Studies 22 (6) (2008) 2201–2238

    2008

  8. [8]

    Zhang, J

    J. Zhang, J. Wang, Modeling and simulation of the market fluctuations by the finite range contact systems, Simulation Modelling Practice and Theory 18 (6) (2010) 910–925

    2010

  9. [9]

    J. Li, L. Cheng, X. Zheng, F.-Y. Wang, Analyzing the stock volatility spillovers in chinese financial and economic sectors, IEEE Transactions on Computational Social Systems 10 (1) (2023) 269–284

    2023

  10. [10]

    A. G. Haldane, R. M. May, Systemic risk in banking ecosystems, Nature 469 (7330) (2011) 351–355

    2011

  11. [11]

    G. W. Imbens, Causal inference in the social sciences, Annual Review of Statistics and Its Application 11 (Volume 11, 2024) (2024) 123–152

    2024

  12. [12]

    Kmenta, Mastering ‘metrics’: The path from cause to effect, Business Economics 50 (4) (2015) 230–231

    J. Kmenta, Mastering ‘metrics’: The path from cause to effect, Business Economics 50 (4) (2015) 230–231

    2015

  13. [13]

    S. D. Campbell, A review of backtesting and backtesting procedures, Finance and Economics Discussion Series 2005-21, Board of Governors of the Federal Reserve System (U.S.) (2005)

    2005

  14. [14]

    K. Luo, N. Jin, J. Ma, Concentrated liquidity in ethereum blockchain’s digital asset trading: Insights from innovative back-testing algorithms, Computational Economics 66 (5) (2025) 3607–3635. Zhong et al.:Preprint submitted to ElsevierPage 18 of 20 EvoMarket

    2025

  15. [15]

    X. Xue, F. Chen, D. Zhou, X. Wang, M. Lu, F.-Y. Wang, Computational experiments for complex social systems—part i: The customization of computational model, IEEE Transactions on Computational Social Systems 9 (5) (2022) 1330–1344

    2022

  16. [16]

    M. D. Gould, M. A. Porter, S. Williams, M. McDonald, D. J. Fenn, S. D. Howison, Limit order books, Quantitative Finance 13 (11) (2013) 1709–1742

    2013

  17. [17]

    X. Xue, D. Zhou, X. Yu, G. Wang, J. Li, X. Xie, L. Cui, F.-Y. Wang, Computational experiments for complex social systems: Experiment design and generative explanation, IEEE/CAA Journal of Automatica Sinica 11 (4) (2024) 1022–1038

    2024

  18. [18]

    B. M. G, P. K. R, V. J. D. V, P. R, V. Maniappan, S. Doss, Enhancing algorithmic trading strategies with sentiment analysis: A reinforcement learning approach, in: 2024 IEEE 3rd World Conference on Applied Intelligence and Computing (AIC), 2024, pp. 107–112

    2024

  19. [19]

    Charles Schwab & Co., Paper trading (thinkorswim papermoney), Web page, accessed: 2026-01-11 (2023)

    2026

  20. [20]

    Nasdaq, Nasdaq Test Facility (NTF) Guide, version 1.3.1 (Dec. 2018)

    2018

  21. [21]

    Hendershott, M

    T. Hendershott, M. Wee, Y. Wen, Transparency in fragmented markets: Experimental evidence, Journal of Financial Markets 59 (2022) 100732

    2022

  22. [22]

    T. H. Balch, M. Mahfouz, J. Lockhart, M. Hybinette, D. Byrd, How to evaluate trading strategies: Single agent market replay or multiple agent interactive simulation? (2019)

    2019

  23. [23]

    Bailey, J

    D. Bailey, J. Borwein, M. Lopez de Prado, Q. J. Zhu, The probability of backtest overfitting, The Journal of Computational Finance 20 (4) (2017) 39–69

    2017

  24. [24]

    D. Byrd, M. Hybinette, T. H. Balch, Abides: Towards high-fidelity multi-agent market simulation, in: Proceedings of the 2020 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation, SIGSIM-PADS ’20, Association for Computing Machinery, New York, NY, USA, 2020, p. 11–22

    2020

  25. [25]

    Belcak, J.-P

    P. Belcak, J.-P. Calliess, S. Zohren, Fast agent-based simulation framework with applications to reinforcement learning and the study of trading latency effects, in: K. H. Van Dam, N. Verstaevel (Eds.), Multi-Agent-Based Simulation XXII, Springer International Publishing, Cham, 2022, pp. 42–56

    2022

  26. [26]

    S. Y. Frey, K. Li, P. Nagy, S. Sapora, C. Lu, S. Zohren, J. Foerster, A. Calinescu, Jax-lob: A gpu-accelerated limit order book simulator to unlock large scale reinforcement learning for trading, in: Proceedings of the Fourth ACM International Conference on AI in Finance, ICAIF ’23, Association for Computing Machinery, New York, NY, USA, 2023, p. 583–591

    2023

  27. [27]

    Abergel, M

    F. Abergel, M. Anane, A. Chakraborti, A. Jedidi, I. M. Toke, Limit Order Books, Cambridge University Press, Cambridge, UK, 2016

    2016

  28. [28]

    LeBaron, Agent-based financial markets: Matching stylized facts with style, Post Walrasian Macroeconomics: Beyond the DSGE Model 221 (2006) 235

    B. LeBaron, Agent-based financial markets: Matching stylized facts with style, Post Walrasian Macroeconomics: Beyond the DSGE Model 221 (2006) 235

    2006

  29. [29]

    Goosen, Calibrating high frequency trading data to agent based models using approximate bayesian computation (2021)

    K. Goosen, Calibrating high frequency trading data to agent based models using approximate bayesian computation (2021)

    2021

  30. [30]

    J. Dyer, P. Cannon, J. D. Farmer, S. M. Schmon, Black-box bayesian inference for agent-based models, Journal of Economic Dynamics and Control 161 (2024) 104827

    2024

  31. [31]

    Platt, A comparison of economic agent-based model calibration methods, Journal of Economic Dynamics and Control 113 (2020) 103859

    D. Platt, A comparison of economic agent-based model calibration methods, Journal of Economic Dynamics and Control 113 (2020) 103859

    2020

  32. [32]

    M. Lu, S. Chen, X. Xue, X. Wang, Y. Zhang, Y. Zhang, F.-Y. Wang, Computational experiments for complex social systems—part ii: The evaluation of computational models, IEEE Transactions on Computational Social Systems 9 (4) (2022) 1224–1236

    2022

  33. [33]

    X. Xue, X. Yu, D. Zhou, C. Peng, X. Wang, D. Liu, F.-Y. Wang, Computational experiments for complex social systems—part iii: The docking of domain models, IEEE Transactions on Computational Social Systems 11 (2) (2024) 1766–1780

    2024

  34. [34]

    Ehrentreich, Agent-based modeling: The Santa Fe Institute artificial stock market model revisited, Springer, 2008

    N. Ehrentreich, Agent-based modeling: The Santa Fe Institute artificial stock market model revisited, Springer, 2008

    2008

  35. [35]

    W. B. Arthur, J. H. Holland, B. LeBaron, R. Palmer, P. Tayler, Asset pricing under endogenous expectations in an artificial stock market, in: The economy as an evolving complex system II, CRC Press, 2018, pp. 15–44

    2018

  36. [36]

    Sagwal, P

    S. Sagwal, P. Kayal, K. Vemuri, Analyzing herding, stylized facts, and information cascades via self-organized criticality in an agent-based speculation game, Simulation Modelling Practice and Theory 144 (2025) 103190.doi:https://doi.org/10.1016/j.simpat.2025.10 3190. URLhttps://www.sciencedirect.com/science/article/pii/S1569190X2500125X

    work page doi:10.1016/j.simpat.2025.10 2025

  37. [37]

    Mascioli, A

    C. Mascioli, A. Gu, Y. Wang, M. Chakraborty, M. Wellman, A financial market simulation environment for trading agents using deep reinforcement learning, in: Proceedings of the 5th ACM International Conference on AI in Finance, ICAIF ’24, Association for Computing Machinery, New York, NY, USA, 2024, p. 117–125

    2024

  38. [38]

    Budish, P

    E. Budish, P. Cramton, J. Shim, The high-frequency trading arms race: Frequent batch auctions as a market design response *, The Quarterly Journal of Economics 130 (4) (2015) 1547–1621

    2015

  39. [39]

    Bogousslavsky, D

    V. Bogousslavsky, D. Muravyev, Who trades at the close? implications for price discovery and liquidity, Journal of Financial Markets 66 (2023) 100852

    2023

  40. [40]

    Chen, A.-P

    C.-C. Chen, A.-P. Chen, P.-Y. Yeh, Modeling and simulation of the open-end equity mutual fund market in taiwan by using self-organizing map, Simulation Modelling Practice and Theory 36 (2013) 60–73

    2013

  41. [41]

    O. U. Aktas, L. Kryzanowski, J. Zhang, Volatility spillover around price limits in an emerging market, Finance Research Letters 39 (2021) 101610

    2021

  42. [42]

    Hautsch, A

    N. Hautsch, A. Horvath, How effective are trading pauses?, Journal of Financial Economics 131 (2) (2019) 378–403

    2019

  43. [43]

    Bongaerts, S

    D. Bongaerts, S. D. De Luca, M. Van Achter, Circuit breakers and market runs, Review of Finance 28 (6) (2024) 1953–1989

    2024

  44. [44]

    Madhavan, Market microstructure: A survey, Journal of Financial Markets 3 (3) (2000) 205–258

    A. Madhavan, Market microstructure: A survey, Journal of Financial Markets 3 (3) (2000) 205–258

    2000

  45. [45]

    R. Cont, M. Cucuringu, C. Zhang, Cross-impact of order flow imbalance in equity markets, Quantitative Finance 23 (10) (2023) 1373–1393

    2023

  46. [46]

    H. Ham, D. Ryu, R. I. Webb, The effects of overnight events on daytime trading sessions, International Review of Financial Analysis 83 (2022) 102228

    2022

  47. [47]

    Zhong, Y

    M. Zhong, Y. Lin, P. Yang, Representation learning of limit order book: A comprehensive study and benchmarking (2025)

    2025

  48. [48]

    Zhong et al.:Preprint submitted to ElsevierPage 19 of 20 EvoMarket

    H.Tian,X.Zhang,X.Zheng,Z.Zhang,D.D.Zeng,Graphrepresentationlearningofmultilayerspatial–temporalnetworksforstockpredictions, IEEE Transactions on Computational Social Systems 12 (5) (2025) 2228–2241. Zhong et al.:Preprint submitted to ElsevierPage 19 of 20 EvoMarket

    2025

  49. [49]

    Y.Li,Y.Wu,M.Zhong,S.Liu,P.Yang,Simlob:Learningrepresentationsoflimitorderbookforfinancialmarketsimulation,IEEETransactions on Artificial Intelligence (2025) 1–16

    2025

  50. [50]

    A. V. Contreras, A. Llanes, A. Pérez-Bernabeu, S. Navarro, H. Pérez-Sánchez, J. J. López-Espín, J. M. Cecilia, Enmx: An elastic network model to predict the forex market evolution, Simulation Modelling Practice and Theory 86 (2018) 1–10

    2018

  51. [51]

    Lamperti, A

    F. Lamperti, A. Roventini, A. Sani, Agent-based model calibration using machine learning surrogates, Journal of Economic Dynamics and Control 90 (2018) 366–389

    2018

  52. [52]

    Jiang, Z

    B. Jiang, Z. Yang, C. Wang, M. Zhong, H. Fang, P. Yang, Calibrating agent-based financial markets simulators with pretrainable automatic posterior transformation-based surrogates (2026)

    2026

  53. [53]

    N. R. Stillman, R. Baggott, J. Lyon, J. Zhang, D. Zhu, T. Chen, P. Vytelingum, Deep calibration of market simulations using neural density estimators and embedding networks, in: Proceedings of the Fourth ACM International Conference on AI in Finance, ICAIF ’23, Association for Computing Machinery, New York, NY, USA, 2023, p. 46–54

    2023

  54. [54]

    P. Yang, Z. Yang, B. Jiang, C. Wang, K. Tang, X. Yao, Posterior distribution-assisted evolutionary dynamic optimization as an online calibrator for complex social simulations (2026)

    2026

  55. [55]

    C. Wang, J. Ren, P. Yang, Alleviating nonidentifiability: A high-fidelity calibration objective for financial market simulation with multivariate time series data, IEEE Transactions on Computational Social Systems 12 (6) (2025) 4910–4922

    2025

  56. [56]

    Cranmer, J

    K. Cranmer, J. Brehmer, G. Louppe, The frontier of simulation-based inference, Proceedings of the National Academy of Sciences 117 (48) (2020) 30055–30062

    2020

  57. [57]

    H.Fang,B.Li,P.Yang,Efficientparametercalibrationofnumericalweatherpredictionmodelsviaevolutionarysequentialtransferoptimization (2026)

    2026

  58. [58]

    R. M. Fujimoto, Parallel discrete event simulation, Commun. ACM 33 (10) (1990) 30–53

    1990

  59. [59]

    Jagtap, N

    D. Jagtap, N. Abu-Ghazaleh, D. Ponomarev, Optimization of parallel discrete event simulator for multi-core systems, in: 2012 IEEE 26th International Parallel and Distributed Processing Symposium, 2012, pp. 520–531

    2012

  60. [60]

    Richmond, R

    P. Richmond, R. Chisholm, P. Heywood, M. K. Chimeh, M. Leach, Flame gpu 2: A framework for flexible and performant agent based simulation on gpus, Software: Practice and Experience 53 (8) (2023) 1659–1680

    2023

  61. [61]

    Samanidou, E

    E. Samanidou, E. Zschischang, D. Stauffer, T. Lux, Agent-based models of financial markets, Reports on Progress in Physics 70 (3) (2007) 409. Zhong et al.:Preprint submitted to ElsevierPage 20 of 20

    2007