<?xml version='1.0' encoding='utf-8'?>
<article xmlns:ns0="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="research-article" dtd-version="1.2" xml:lang="ru">
 <front>
 <journal-meta>
 <journal-id journal-id-type="publisher-id">dongu-nec.ru</journal-id>
 <journal-title-group>
 <journal-title xml:lang="ru">Новое в экономической кибернетике</journal-title>
 <trans-title-group xml:lang="en">
 <trans-title>New in Economic Cybernetics</trans-title>
 </trans-title-group>
 </journal-title-group>
 <issn publication-format="electronic">2523-448X</issn>
 </journal-meta>
 <article-meta>
 <article-id pub-id-type="publisher-id">531</article-id>
 <article-id pub-id-type="doi">10.5281/zenodo.20970581</article-id>
 <article-categories>
 <subj-group subj-group-type="toc-heading" xml:lang="en">
 <subject>Articles</subject>
 </subj-group>
 <subj-group subj-group-type="toc-heading" xml:lang="ru">
 <subject>Статьи</subject>
 </subj-group>
 <subj-group subj-group-type="article-type">
 <subject>Research Article</subject>
 </subj-group>
 </article-categories>
 <title-group>
 <article-title xml:lang="en">NETWORK-BASED METHODS OF PORTFOLIO OPTIMIZATION: MAXIMUM INDEPENDENT SETS AND GRAPH ATTENTION NETWORKS IN THE RUSSIAN STOCK MARKET</article-title>
 <trans-title-group xml:lang="ru">
 <trans-title>СЕТЕВЫЕ МЕТОДЫ ПОРТФЕЛЬНОЙ ОПТИМИЗАЦИИ: МАКСИМАЛЬНЫЕ НЕЗАВИСИМЫЕ МНОЖЕСТВА И ГРАФОВЫЕ НЕЙРОСЕТИ ВНИМАНИЯ НА РЫНКЕ АКЦИЙ РОССИИ</trans-title>
 </trans-title-group>
 </title-group>
 <contrib-group>
 <contrib contrib-type="author">
 <contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-3963-0814</contrib-id>
 <name-alternatives>
 <name xml:lang="en">
 <surname>Kulikov</surname>
 <given-names>Alexander</given-names>
 </name>
 <name xml:lang="ru">
 <surname>Куликов</surname>
 <given-names>Александр Владимирович</given-names>
 </name>
 </name-alternatives>
 <email>kulikov.alexandr@phystech.edu</email>
 <xref ref-type="aff" rid="aff1">1</xref>
 </contrib>
 <contrib contrib-type="author">
 <contrib-id contrib-id-type="orcid">https://orcid.org/0009-0002-1601-4856</contrib-id>
 <name-alternatives>
 <name xml:lang="en">
 <surname>Panagiotou</surname>
 <given-names>Elina</given-names>
 </name>
 <name xml:lang="ru">
 <surname>Панайоту</surname>
 <given-names>Элина Михайловна</given-names>
 </name>
 </name-alternatives>
 <email>panaiotu.em@phystech.edu</email>
 <xref ref-type="aff" rid="aff2">2</xref>
 </contrib>
 </contrib-group>
 <aff-alternatives id="aff1">
 <aff xml:lang="ru">
 <institution>ФГАОУ ВО «Московский физико-технический институт (национальный исследовательский университет)»</institution>
 </aff>
 <aff xml:lang="en">
 <institution>Moscow Institute of Physics and Technology (National Research University)</institution>
 </aff>
 </aff-alternatives>
 <aff-alternatives id="aff2">
 <aff xml:lang="ru">
 <institution>ФГАОУ ВО «Московский физико-технический институт (национальный исследовательский университет)»</institution>
 </aff>
 <aff xml:lang="en">
 <institution>Moscow Institute of Physics and Technology (National Research University)</institution>
 </aff>
 </aff-alternatives>
 <pub-date date-type="pub" iso-8601-date="2026-07-01" publication-format="electronic">
 <day>01</day>
 <month>07</month>
 <year>2026</year>
 </pub-date>
 <issue>2</issue>
 <issue-title xml:lang="en">NO2 (2026)</issue-title>
 <issue-title xml:lang="ru">№2 (2026)</issue-title>
 <fpage>79</fpage>
 <lpage>94</lpage>
 <history>
 <date date-type="received" iso-8601-date="2026-07-07">
 <day>07</day>
 <month>07</month>
 <year>2026</year>
 </date>
 </history>
 <permissions>
 <copyright-statement xml:lang="ru">Copyright ©; 2026, Новое в экономической кибернетике</copyright-statement>
 <copyright-statement xml:lang="en">Copyright ©; 2026, New in Economic Cybernetics</copyright-statement>
 <copyright-year>2026</copyright-year>
 <copyright-holder xml:lang="ru">Новое в экономической кибернетике</copyright-holder>
 <copyright-holder xml:lang="en">New in Economic Cybernetics</copyright-holder>
 <license license-type="open-access" ns0:href="https://creativecommons.org/licenses/by-nc/4.0/" xml:lang="ru">
 <license-p>Эта статья распространяется на условиях лицензии Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)</license-p>
 </license>
 <license license-type="open-access" ns0:href="https://creativecommons.org/licenses/by-nc/4.0/" xml:lang="en">
 <license-p>This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0)</license-p>
 </license>
 <ali:free_to_read />
 </permissions>
 <self-uri ns0:href="https://dongu-nec.ru/index.php/nec/article/view/531">https://dongu-nec.ru/index.php/nec/article/view/531</self-uri>
 <abstract xml:lang="en">
 <p>The paper develops and empirically evaluates a portfolio optimization approach based on graph neural networks with an attention mechanism (Graph Attention Network, GAT), combined with a preliminary structural filtering of the asset universe using the Maximum Independent Set (MIS) and an adaptive correlation threshold. In addition, the study provides a concise analysis of of the Russian equity market graph characteristics and the dynamics of maximum independent sets. The empirical dataset consists of liquid Russian equities traded on the MOEX over the period 2020–2025, covering both calm regimes and stress episodes accompanied by structural breaks. The proposed method is compared with the market index, classical portfolio strategies, a multilayer perceptron, and a graph‑based model without MIS filtering within a walk‑forward time‑series validation scheme. The results show that incorporating MIS with an adaptive threshold leads to substantial improvements across all key metrics relative to the baseline GAT model without filtering: on average, final portfolio value increases by about 56%, the Sharpe ratio by about 157%, the maximum drawdown in absolute value decreases by 35%, and the average within‑portfolio correlation falls by 43%. The performance gain is particularly pronounced in down‑market phases, when the baseline GAT yields negative returns. An entropy‑based analysis of the attention coefficients indicates that MIS filtering makes the model’s attention more concentrated, reallocating it towards a small number of structurally important edges and reducing the impact of noisy connections.</p>
 </abstract>
 <trans-abstract xml:lang="ru">
 <p>Статья посвящена разработке и эмпирической проверке подхода к управлению инвестиционным портфелем на основе графовых нейронных сетей c механизмом внимания (Graph attention network, GAT) с предварительной структурной фильтрацией множества активов с использованием максимального независимого множества (Maximum Independent Set, MIS) и адаптивного порога корреляции. Помимо этого, работа содержит краткий анализ графовых характеристик российского рынка и динамики максимальных независимых множеств. Эмпирическая база исследования включает данные по ликвидным акциям российского фондового рынка за период 2020–2025 годов, охватывающий как относительно спокойные режимы, так и стрессовые эпизоды и структурные сдвиги. Предлагаемый метод сравнивается с рыночным индексом, классическими портфельными стратегиями, многослойным персептроном и графовой моделью без MIS‑фильтрации с использованием walk‑forward оптимизации. Результаты показывают, что включение MIS с адаптивным порогом приводит к заметному улучшению по всем ключевым метрикам относительно базовой GAT‑модели без фильтрации. Итоговый капитал в среднем возрастает примерно на 56 %, коэффициент Шарпа – примерно на 157 %, максимальная просадка по модулю уменьшается на 35 %, а средняя внутрипортфельная корреляция снижается на 43 %. При этом на падающем рынке, когда базовый GAT даёт отрицательную доходность, выигрыш стратегии особенно заметен. Анализ энтропии коэффициентов внимания свидетельствует о том, что MIS‑фильтрация делает внимание модели более сконцентрированным, перераспределяя его в пользу малого числа структурно значимых рёбер и снижая влияние шумовых связей.</p>
 </trans-abstract>
 <kwd-group xml:lang="en">
 <kwd>diversification, portfolio optimization, Russian stock market, dependency network, maximum independent set, graph attention networks, Sharpe ratio.</kwd>
 </kwd-group>
 <kwd-group xml:lang="ru">
 <kwd>диверсификация, портфельная оптимизация, фондовый рынок России, граф взаимосвязей, максимальное независимое множество, графовые нейросети внимания, коэффициент Шарпа.</kwd>
 </kwd-group>
 </article-meta>
 </front>
 <body>
 <p>[Полный текст статьи отсутствует в исходных данных. Необходимо добавить текст из PDF или другого источника.]</p>
 </body>
 <back>
 <ref-list>
 <title>Список литературы</title>
 <ref id="B1">
 <mixed-citation>1. Markowitz, H. Portfolio Selection // The Journal of Finance. 1952. Vol. 7, no. 1. P. 77–91.</mixed-citation>
 </ref>
 <ref id="B2">
 <mixed-citation>2. Sharpe, W.F. Mutual Fund Performance // The Journal of Business. 1966. Vol. 39, no. S1. P. 119–138.</mixed-citation>
 </ref>
 <ref id="B3">
 <mixed-citation>3. Engle, R. Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models // Journal of Business &amp;Economic Statistics. 2002. Vol. 20, no. 3. P. 339–350.</mixed-citation>
 </ref>
 <ref id="B4">
 <mixed-citation>4. Jagannathan Ravi, Ma Tongshu. Risk Reduction in Large Portfolios: Why Imposing the Wrong Constraints Helps // The Journal of Finance. 2003. Т. 58, № 4. P. 1651–1684.</mixed-citation>
 </ref>
 <ref id="B5">
 <mixed-citation>5. Mantegna, R.N. Hierarchical Structure in Financial Markets // The European Physical Journal B. 1999. Vol. 11, no. 1. P. 193–197</mixed-citation>
 </ref>
 <ref id="B6">
 <mixed-citation>6. A Tool for Filtering Information in Complex Systems / M. Tumminello, T. Aste, T. Di Matteo et al. // Proceedings of the National Academy of Sciences. 2005. Vol.102, no. 30. P. 10421–10426.</mixed-citation>
 </ref>
 <ref id="B7">
 <mixed-citation>7. Millington T., Niranjan M. Partial Correlation Financial Networks // Applied Network Science. 2020. Vol. 5, no. 1. P. 11.</mixed-citation>
 </ref>
 <ref id="B8">
 <mixed-citation>8. Dominating Clasp of the Financial Sector Revealed by Partial Correlation Analysis of the Stock Market / D. Y. Kenett, M. Tumminello, A. Madi et al. // PLoS ONE. 2010. Vol. 5, no. 12. P. e15032.</mixed-citation>
 </ref>
 <ref id="B9">
 <mixed-citation>9. Correlation-Diversified Portfolio Construction by Finding Maximum Independent Set in Large-Scale Market Graph / R. Hidaka, Y. Hamakawa, J. Nakayama et al. // IEEE Access. 2023. Vol. 11. P. 142979–142991.</mixed-citation>
 </ref>
 <ref id="B10">
 <mixed-citation>10. Veliˇckovi´c P., Cucurull G., Casanova A. et al. Graph Attention Networks. Preprint on arXiv. 2018.</mixed-citation>
 </ref>
 <ref id="B11">
 <mixed-citation>11. A Comprehensive Survey on Graph Neural Networks / Z. Wu, S. Pan, F. Chen et al. // IEEE Transactions on Neural Networks and Learning Systems. 2021. Vol. 32, no. 1. P. 4 24.</mixed-citation>
 </ref>
 <ref id="B12">
 <mixed-citation>12. Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction / S. Xiang, D. Cheng, C. Shang et al. // Proceedings of the 31st ACM International Conference on Information and Knowledge Management. 2022. P. 3584–3593.</mixed-citation>
 </ref>
 <ref id="B13">
 <mixed-citation>13. Jang J., Oh D., Park S. Combining Graph Attention Networks and Modern Portfolio Theory for Enhanced Stock Portfolio Optimization. Preprint on SSRN. 2024.</mixed-citation>
 </ref>
 <ref id="B14">
 <mixed-citation>14. Визгунов, А.Н. Применение рыночных графов к анализу фондового рынка России / А.Н. Визгунов, Б.И. Гольденгорин, В.А. Замараев [и др.] // Журнал Новой экономической ассоциации. – 2012. – № 3(15). – С. 66–81.</mixed-citation>
 </ref>
 <ref id="B15">
 <mixed-citation>15. Comparative Analysis of Financial Network Topology for the Russian, Chinese and US Stock Markets / V. Balash, S. Sidorov, A. Faizliev et al. // WSEAS Transactions on Business and Economics. 2020. Vol. 17. P. 120–132.</mixed-citation>
 </ref>
 <ref id="B16">
 <mixed-citation>16. Eratalay M. H., Vladimirov E. Mapping the Stocks in MICEX: Who Is Central to the Moscow Stock Exchange? // SSRN Electronic Journal. 2018.</mixed-citation>
 </ref>
 <ref id="B17">
 <mixed-citation>17. Shternshis A., Mazzarisi P., Marmi S. Efficiency of the Moscow Stock Exchange Before 2022 // Entropy. 2022. Vol. 24, no. 9. P. 1184.</mixed-citation>
 </ref>
 <ref id="B18">
 <mixed-citation>18. Гребенников, Н.Э. Применение алгоритмов машинного обучения для оптимизации портфеля ценных бумаг // Управленческий учет. – 2025. – № 2. – С. 240–249.</mixed-citation>
 </ref>
 <ref id="B19">
 <mixed-citation>19. РБК. Топ-5 потрясений инвесторов на российском рынке в 2022 году // РБК Инвестиции. – 2023. – URL: https://www.rbc.ru/quote/news/article/ 63a9aac99a79476688cf5ed6 (дата обращения: 25.04.2026).</mixed-citation>
 </ref>
 <ref id="B20">
 <mixed-citation>20. Forbes Россия. «Индекс страха» российского рынка акций в 2025 году достиг максимума за 10 лет // Forbes Россия. 2026. – URL: https://www.forbes.ru/investicii/553842-indeks-straha-rossijskogo-rynka-akcij-v-2025-godu-dostig-maksimuma-za-10-let (дата обращения: 25.04.2026).</mixed-citation>
 </ref>
 <ref id="B21">
 <mixed-citation>21. Московская Биржа. Частные инвесторы в 2025 году вложили в ценные бумаги на Мосбирже рекордные 2,5 трлн рублей // Московская Биржа. – 2026. – URL: https://www.moex.com/n96827 (дата обращения: 25.04.2026).</mixed-citation>
 </ref>
 <ref id="B22">
 <mixed-citation>22. Альфа-Банк. Результаты народного портфеля в 2025 году // Альфа-Банк. – 2026. – URL: https://alfabank.ru/alfa-investor/t/rezultaty-narodnogo-portfelya-v-2025-godu/ (дата обращения: 25.04.2026)</mixed-citation>
 </ref>
 <ref id="B23">
 <mixed-citation>References</mixed-citation>
 </ref>
 <ref id="B24">
 <mixed-citation>1. Markowitz H. (1952) Portfolio Selection. The Journal of Finance. Vol. 7, no. 1. Pp. 77–91.</mixed-citation>
 </ref>
 <ref id="B25">
 <mixed-citation>2. Sharpe, W.F. (1966) Mutual Fund Performance. The Journal of Business. Vol. 39, no. S1. Pp. 119–138.</mixed-citation>
 </ref>
 <ref id="B26">
 <mixed-citation>3. Engle, R. (2002) Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models. Journal of Business &amp; Economic Statistics. Vol. 20, no. 3. Pp. 339–350.</mixed-citation>
 </ref>
 <ref id="B27">
 <mixed-citation>4. Jagannathan R., Ma T. (2003) Risk Reduction in Large Portfolios: Why Imposing the Wrong Constraints Helps. The Journal of Finance. Vol. 58, no. 4. Pp. 1651–1684.</mixed-citation>
 </ref>
 <ref id="B28">
 <mixed-citation>5. Mantegna, R.N. (1999) Hierarchical Structure in Financial Markets. The European Physical Journal B. Vol. 11, no. 1. Pp. 193–197.</mixed-citation>
 </ref>
 <ref id="B29">
 <mixed-citation>6. Tumminello M., Aste T., Di Matteo T. et al. (2005) A Tool for Filtering Information in Complex Systems. Proceedings of the National Academy of Sciences. Vol. 102, no. 30. Pp. 10421–10426.</mixed-citation>
 </ref>
 <ref id="B30">
 <mixed-citation>7. Millington T., Niranjan M. (2020) Partial Correlation Financial Networks. Applied Network Science. Vol. 5, no. 1. Article 11.</mixed-citation>
 </ref>
 <ref id="B31">
 <mixed-citation>8. Kenett D.Y., Tumminello M., Madi A. et al. (2010) Dominating Clasp of the Financial Sector Revealed by Partial Correlation Analysis of the Stock Market. PLoS ONE. Vol. 5, no. 12. e15032.</mixed-citation>
 </ref>
 <ref id="B32">
 <mixed-citation>9. Hidaka R., Hamakawa Y., Nakayama J. et al. (2023) Correlation-Diversified Portfolio Construction by Finding Maximum Independent Set in Large-Scale Market Graph. IEEE Access. Vol. 11. Pp. 142979–142991.</mixed-citation>
 </ref>
 <ref id="B33">
 <mixed-citation>10. Veličković P., Cucurull G., Casanova A. et al. (2018) Graph Attention Networks. arXiv preprint. arXiv:1710.10903.</mixed-citation>
 </ref>
 <ref id="B34">
 <mixed-citation>11. Wu Z., Pan S., Chen F. et al. (2021) A Comprehensive Survey on Graph Neural Networks. IEEE Transactions on Neural Networks and Learning Systems. Vol. 32, no. 1. Pp. 4 24.</mixed-citation>
 </ref>
 <ref id="B35">
 <mixed-citation>12. Xiang S., Cheng D., Shang C. et al. (2022) Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction. Proceedings of the 31st ACM International Conference on Information and Knowledge Management. Pp. 3584–3593.</mixed-citation>
 </ref>
 <ref id="B36">
 <mixed-citation>13. Jang J., Oh D., Park S. (2024) Combining Graph Attention Networks and Modern Portfolio Theory for Enhanced Stock Portfolio Optimization. SSRN preprint.</mixed-citation>
 </ref>
 <ref id="B37">
 <mixed-citation>14. Vizgunov A.N., Goldengorin B.I., Zamaraev V.A. et al. (2012) Application of Market Graphs to the Analysis of the Russian Stock Market. Journal of the New Economic Association. No. 3(15). Pp. 66–81. (In Russian).</mixed-citation>
 </ref>
 <ref id="B38">
 <mixed-citation>15. Balash V., Sidorov S., Faizliev A. et al. (2020) Comparative Analysis of Financial Network Topology for the Russian, Chinese and US Stock Markets. WSEAS Transactions on Business and Economics. Vol. 17. Pp. 120–132.</mixed-citation>
 </ref>
 <ref id="B39">
 <mixed-citation>16. Eratalay M.H., Vladimirov E. (2018) Mapping the Stocks in MICEX: Who Is Central to the Moscow Stock Exchange? SSRN Electronic Journal.</mixed-citation>
 </ref>
 <ref id="B40">
 <mixed-citation>17. Shternshis A., Mazzarisi P., Marmi S. (2022) Efficiency of the Moscow Stock Exchange Before 2022. Entropy. Vol. 24, no. 9. Article 1184.</mixed-citation>
 </ref>
 <ref id="B41">
 <mixed-citation>18. Grebennikov, N.E. (2025) Application of Machine Learning Algorithms to the Optimization of a Securities Portfolio. Management Accounting. No. 2. Pp. 240–249. (In Russian).</mixed-citation>
 </ref>
 <ref id="B42">
 <mixed-citation>19. RBC. Top 5 Shocks for Investors on the Russian Market in 2022 // RBC Investments. 2023. URL: https://www.rbc.ru/quote/news/article/63a9aac99a79476688cf5ed6 (accessed: 25.04.2026). (In Russian).</mixed-citation>
 </ref>
 <ref id="B43">
 <mixed-citation>20. Forbes Russia. The “Fear Index” of the Russian Equity Market in 2025 Reached Its 10 Year High // Forbes Russia. 2026. URL: https://www.forbes.ru/investicii/553842-indeks-straha-rossijskogo-rynka-akcij-v-2025-godu-dostig-maksimuma-za-10-let (accessed: 25.04.2026). (In Russian).</mixed-citation>
 </ref>
 <ref id="B44">
 <mixed-citation>21. Moscow Exchange. Retail Investors Invested a Record 2.5 Trillion Rubles in Securities on the Moscow Exchange in 2025 // Moscow Exchange. 2026. URL: https://www.moex.com/n96827 (accessed: 25.04.2026). (In Russian).</mixed-citation>
 </ref>
 <ref id="B45">
 <mixed-citation>22. Alfa Bank. Performance of the “People’s Portfolio” in 2025 // Alfa Bank. 2026. URL: https://alfabank.ru/alfa-investor/t/rezultaty-narodnogo-portfelya-v-2025-godu/ (accessed: 25.04.2026). (In Russian).</mixed-citation>
 </ref>
 </ref-list>
 </back>
 </article>
