IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2607.23068.html

Neural Network-Driven Volatility Drag Mitigation under Aggressive Leverage

Author

Listed:
  • Christian Bongiorno
  • Efstratios Manolakis
  • Rosario Nunzio Mantegna

Abstract

This paper introduces a compact reformulation of a modular end-to-end neural network for global minimum-variance portfolio optimization that decouples model complexity from both look-back window length and universe size. A five-parameter hyperbolic weighted moving average combined with a saturating exponential replaces the original 2,400-parameter lag-transformation layer, and a bidirectional gated-recurrent-unit eigencleaning module together with a streamlined marginal-volatility network reduce total learnable parameters from 39,586 to just 2,175. In out-of-sample tests against state-of-the-art nonlinear-shrinkage and risk-parity benchmarks, the compact network attains the lowest realized portfolio variance without compromising expected return. Under long-only constraints, the variance reduction supports substantially higher leverage while maintaining comparable drawdown control. Validation in a high-fidelity trading simulator that incorporates realistic margin-call dynamics confirms enhanced over-leverage resilience. These findings demonstrate that end-to-end variance-minimization architectures can achieve substantial parameter efficiency and robust capital-efficiency gains without sacrificing risk-adjusted performance.

Suggested Citation

  • Christian Bongiorno & Efstratios Manolakis & Rosario Nunzio Mantegna, 2026. "Neural Network-Driven Volatility Drag Mitigation under Aggressive Leverage," Papers 2607.23068, arXiv.org.
    • Handle: RePEc:arx:papers:2607.23068
    as

    Download full text from publisher

    File URL: https://arxiv.org/pdf/2607.23068
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Robert F. Engle & Olivier Ledoit & Michael Wolf, 2019. "Large Dynamic Covariance Matrices," Journal of Business & Economic Statistics, Taylor & Francis Journals, vol. 37(2), pages 363-375, April.
    2. Mohammed Salek & Damien Challet & Ioane Muni Toke, 2023. "Price impact in equity auctions: zero, then linear," Papers 2301.05677, arXiv.org, revised Sep 2023.
    3. repec:dau:papers:123456789/4688 is not listed on IDEAS
    4. repec:hal:wpaper:hal-03938660 is not listed on IDEAS
    5. Mark Rubinstein, 2002. "Markowitz's “Portfolio Selection”: A Fifty‐Year Retrospective," Journal of Finance, American Finance Association, vol. 57(3), pages 1041-1045, June.
    6. Lorenzo Garlappi & Raman Uppal & Tan Wang, 2007. "Portfolio Selection with Parameter and Model Uncertainty: A Multi-Prior Approach," The Review of Financial Studies, Society for Financial Studies, vol. 20(1), pages 41-81, January.
    7. Christian Bongiorno & Damien Challet & Grégoire Loeper, 2023. "Filtering time-dependent covariance matrices using time-independent eigenvalues," Post-Print hal-03481441, HAL.
    8. Bruce Jacobs & Kenneth Levy, 2014. "The unique risks of portfolio leverage: why modern portfolio theory fails and how to fix it," Journal of Financial Perspectives, EY Global FS Institute, vol. 2(3), pages 113-126.
    9. Peter R. Hansen & Asger Lunde & James M. Nason, 2011. "The Model Confidence Set," Econometrica, Econometric Society, vol. 79(2), pages 453-497, March.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Jean-David Fermanian & Benjamin Poignard & Panos Xidonas, 2025. "Model-based vs. agnostic methods for the prediction of time-varying covariance matrices," Annals of Operations Research, Springer, vol. 346(1), pages 511-548, March.
    2. JD Opdyke, 2025. "Causal Discovery via Simultaneous DAG Recovery Using the Angles Space of Directional Dependence Measures," Papers 2504.15268, arXiv.org, revised Aug 2026.
    3. Golosnoy, Vasyl & Gribisch, Bastian, 2022. "Modeling and forecasting realized portfolio weights," Journal of Banking & Finance, Elsevier, vol. 138(C).
    4. Paul Ruelloux & Christian Bongiorno & Damien Challet, 2025. "Noise-proofing Universal Portfolio Shrinkage," Papers 2511.10478, arXiv.org.
    5. Grønborg, Niels S. & Lunde, Asger & Olesen, Kasper V. & Vander Elst, Harry, 2022. "Realizing correlations across asset classes," Journal of Financial Markets, Elsevier, vol. 59(PA).
    6. Niels S. Grønborg & Asger Lunde & Kasper V. Olesen & Harry Vander Elst, 2018. "Realizing Correlations Across Asset Classes," CREATES Research Papers 2018-37, Department of Economics and Business Economics, Aarhus University.
    7. Kolm, Petter N. & Tütüncü, Reha & Fabozzi, Frank J., 2014. "60 Years of portfolio optimization: Practical challenges and current trends," European Journal of Operational Research, Elsevier, vol. 234(2), pages 356-371.
    8. Marc Hallin & Carlos Trucíos, 2020. "Forecasting Value-at-Risk and Expected Shortfall in Large Portfolios: a General Dynamic Factor Approach," Working Papers ECARES 2020-50, ULB -- Universite Libre de Bruxelles.
    9. Hallin, Marc & Trucíos, Carlos, 2023. "Forecasting value-at-risk and expected shortfall in large portfolios: A general dynamic factor model approach," Econometrics and Statistics, Elsevier, vol. 27(C), pages 1-15.
    10. Alessandra Amendola & Vincenzo Candila & Antonio Naimoli & Giuseppe Storti, 2024. "Combining Value-at-Risk and Expected Shortfall forecasts via the Model Confidence Set," Papers 2406.06235, arXiv.org, revised Feb 2026.
    11. Benjamin Poignard & Manabu Asai, 2023. "High‐dimensional sparse multivariate stochastic volatility models," Journal of Time Series Analysis, Wiley Blackwell, vol. 44(1), pages 4-22, January.
    12. Emilija Dzuverovic & Matteo Barigozzi, 2023. "Hierarchical DCC-HEAVY Model for High-Dimensional Covariance Matrices," Papers 2305.08488, arXiv.org, revised Jul 2024.
    13. Jan Patrick Hartkopf, 2023. "Composite forecasting of vast-dimensional realized covariance matrices using factor state-space models," Empirical Economics, Springer, vol. 64(1), pages 393-436, January.
    14. Christian Bongiorno & Efstratios Manolakis & Rosario Nunzio Mantegna, 2025. "End-to-End Large Portfolio Optimization for Variance Minimization with Neural Networks through Covariance Cleaning," Papers 2507.01918, arXiv.org, revised Apr 2026.
    15. Ahmed, Shamim & Bu, Ziwen & Symeonidis, Lazaros & Tsvetanov, Daniel, 2023. "Which factor model? A systematic return covariation perspective," Journal of International Money and Finance, Elsevier, vol. 136(C).
    16. Hautsch, Nikolaus & Voigt, Stefan, 2019. "Large-scale portfolio allocation under transaction costs and model uncertainty," Journal of Econometrics, Elsevier, vol. 212(1), pages 221-240.
    17. Archakov, Ilya & Hansen, Peter Reinhard & Lunde, Asger, 2026. "A multivariate realized GARCH model," Journal of Econometrics, Elsevier, vol. 254(PA).
    18. Bollerslev, Tim & Patton, Andrew J. & Quaedvlieg, Rogier, 2020. "Multivariate leverage effects and realized semicovariance GARCH models," Journal of Econometrics, Elsevier, vol. 217(2), pages 411-430.
    19. Khashanah, Khaldoun & Simaan, Majeed & Simaan, Yusif, 2022. "Do we need higher-order comoments to enhance mean-variance portfolios? Evidence from a simplified jump process," International Review of Financial Analysis, Elsevier, vol. 81(C).
    20. Ding, Yi & Li, Yingying & Zheng, Xinghua, 2021. "High dimensional minimum variance portfolio estimation under statistical factor models," Journal of Econometrics, Elsevier, vol. 222(1), pages 502-515.

    More about this item

    NEP fields

    This paper has been announced in the following NEP Reports:

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2607.23068. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: https://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.