<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" 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" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">scires</journal-id><journal-title-group><journal-title xml:lang="ru">Научные исследования экономического факультета. Электронный журнал</journal-title><trans-title-group xml:lang="en"><trans-title>Scientific Research of Faculty of Economics. Electronic Journal</trans-title></trans-title-group></journal-title-group><issn pub-type="epub">2078-3809</issn><publisher><publisher-name>Moscow State University</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.38050/2078-3809-2026-18-2-43-57</article-id><article-id custom-type="elpub" pub-id-type="custom">scires-466</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>МАКРОЭКОНОМИЧЕСКАЯ ПОЛИТИКА</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>MACROECONOMIC POLICY</subject></subj-group></article-categories><title-group><article-title>Оценка предсказательной силы инвесторов в определении ключевой ставки ЦБ РФ с помощью LLM</article-title><trans-title-group xml:lang="en"><trans-title>Assessing the Predictive Power of Investors in Determining the Key Rate of the Central Bank of the Russian Federation Using LLM</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Борисенко</surname><given-names>Г. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Borisenko</surname><given-names>G. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Борисенко Георгий Александрович — аспирант</p></bio><bio xml:lang="en"><p>Georgii A. Borisenko — Postgraduate student</p></bio><email xlink:type="simple">borisenko.georgiy@bk.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>МГУ имени М.В. Ломоносова, экономический факультет</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Lomonosov Moscow State University, Faculty of Economics</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2026</year></pub-date><pub-date pub-type="epub"><day>18</day><month>09</month><year>2026</year></pub-date><volume>18</volume><issue>2</issue><fpage>43</fpage><lpage>57</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Борисенко Г.А., 2026</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="ru">Борисенко Г.А.</copyright-holder><copyright-holder xml:lang="en">Borisenko G.A.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://scires.elpub.ru/jour/article/view/466">https://scires.elpub.ru/jour/article/view/466</self-uri><abstract><p>Исследование направлено на прогнозирование решений ЦБ РФ по ключевой ставке через анализ мнений инвесторов в социальной сети «Пульс» (Т-Банк). Несмотря на важность ключевой ставки ЦБ, большинство исследователей ставят целью прогнозирование макроэкономических индикаторов, связанных со ставкой ЦБ, но не ее саму. Для восполнения этого пробела в работе предложен подход, сочетающий краудсорсинговые данные с обработкой большими языковыми моделями (LLM). На основе семантически размеченных постов и комментариев из социальных сетей с помощью YandexGPT-5-Lite-Instruct, построены прогнозы направления изменения ставки за 48 заседаний ЦБ РФ (2019–2025 гг.). Интеграция LLM позволила корректно интерпретировать контекст и эмоциональную окраску неформальных высказываний, избежав ошибок словарных методов. Точность прогнозов на основе данных «Пульса» составила 90%, превысив показатели аналитических агентств в условиях экстренных заседаний.</p></abstract><trans-abstract xml:lang="en"><p>The main aim of this study is to forecast key rate decisions of the Central Bank of Russian Federation by analyzing investor opinions in the Pulse social network (T-Bank). Despite the importance of the Central Bank's key rate, most studies aim to forecast macroeconomic indicators related to the Central Bank rate, but not the rate itself. To fill this gap, the paper proposes an approach that combines crowdsourcing data with processing by large language models (LLM). Based on semantically tagged posts and comments from the social network using YandexGPT-5-Lite-Instruct, forecasts directions of rate changes for 48 meetings of the Central Bank of the Russian Federation (2019–2025). The integration of LLM made it possible to correctly interpret the context and emotional coloring of informal statements, avoiding the errors of dictionary methods. The accuracy of forecasts based on Pulse data was 90%, exceeding the indicators of analytical agencies in the context of emergency meetings.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>ставка ЦБ РФ</kwd><kwd>прогнозирование</kwd><kwd>большие языковые модели</kwd><kwd>анализ настроений</kwd></kwd-group><kwd-group xml:lang="en"><kwd>Central Bank of the Russian Federation rate</kwd><kwd>forecasting</kwd><kwd>large language models</kwd><kwd>sentiment analysis</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Абдурахманов М.А. Моделирование влияния прогнозов ключевой ставки Банка России на ожидания участников рынка // Деньги и кредит. 2023. Т. 82. № 2. С. 3–20.</mixed-citation><mixed-citation xml:lang="en">Abdurahmanov M.A. Modelirovanie vlijanija prognozov kljuchevoj stavki Banka Rossii na ozhidanija uchastnikov rynka. Den'gi i kredit. 2023. Vol. 82. No. 2. P. 3–20. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Абрамов В., Тишин А.В., Стырин К.А. Денежно-кредитная политика и кривая доходности // Серия докладов об экономических исследованиях. 2022. № 95В. С. 49.</mixed-citation><mixed-citation xml:lang="en">Abramov V., Tishin A.V., Styrin K.A. Denezhno-kreditnaja politika i krivaja dohodnosti. Serija dokladov ob jekonomicheskih issledovanijah. 2022. No.95V. P. 49. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Шарафутдинов А.Р. Прогнозирование российских ВВП, инфляции, ставки процента и обменного курса с помощью модели DSGE-VAR // Деньги и кредит. 2023. Т. 82. № 3. С. 32–86.</mixed-citation><mixed-citation xml:lang="en">Sharafutdinov A.R. Prognozirovanie rossijskih VVP, infljacii, stavki procenta i obmennogo kursa s pomoshh'ju modeli DSGE-VAR. Den'gi i kredit. 2023. Vol. 82. No. 3. P. 32–86. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Юревич М.А., Ахмадеев Д.Р. Возможности прогнозирования уровня безработицы на основе анализа статистики запросов (в поисковых системах) // Terra Economicus. 2021. Т. 19. № 3. С. 53–64. https://doi.org/10.18522/2073-6606-2021-19-3-53-64.</mixed-citation><mixed-citation xml:lang="en">Jurevich M.A., Ahmadeev D.R. Vozmozhnosti prognozirovanija urovnja bezraboticy na osnove analiza statistiki zaprosov (v poiskovyh sistemah). Terra Economicus. 2021. Vol. 19. No. 3. P. 53–64. https://doi.org/10.18522/2073-6606-2021-19-3-53-64. (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Darapaneni N., Paduri A.R., Sharma H. et al. Stock price prediction using sentiment analysis and deep learning for Indian markets // arXiv preprint arXiv:2204.05783. 2022. https://doi.org/10.48550/arXiv.2204.05783.</mixed-citation><mixed-citation xml:lang="en">Darapaneni N., Paduri A.R., Sharma H. et al. Stock price prediction using sentiment analysis and deep learning for Indian markets. arXiv preprint arXiv:2204.05783. 2022. https://doi.org/10.48550/arXiv.2204.05783.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Islam M.R., Zibran M.F. A comparison of dictionary building methods for sentiment analysis in software engineering text // 2017 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM). IEEE, 2017. P. 478–479. https://doi.org/10.1109/ESEM.2017.67.</mixed-citation><mixed-citation xml:lang="en">Islam M.R., Zibran M.F. A comparison of dictionary building methods for sentiment analysis in software engineering text. 2017 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM). IEEE, 2017. P. 478–479. https://doi.org/10.1109/ESEM.2017.67.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Jagdale R.S., Shirsat V.S., Deshmukh S.N. Sentiment analysis on product reviews using machine learning techniques // Cognitive Informatics and Soft Computing: Proceeding of CISC 2017. Springer Singapore, 2019. P. 639–647. https://doi.org/10.1007/978-981-13-0617-4_61.</mixed-citation><mixed-citation xml:lang="en">Jagdale R.S., Shirsat V.S., Deshmukh S.N. Sentiment analysis on product reviews using machine learning techniques. Cognitive Informatics and Soft Computing: Proceeding of CISC 2017. Springer Singapore, 2019. P. 639–647. https://doi.org/10.1007/978-981-13-0617-4_61.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Luo W., Gong D. Pre-trained large language models for financial sentiment analysis // arXiv preprint arXiv:2401.05215. 2024. P. 6. https://doi.org/10.48550/arXiv.2401.05215.</mixed-citation><mixed-citation xml:lang="en">Luo W., Gong D. Pre-trained large language models for financial sentiment analysis. arXiv preprint arXiv:2401.05215. 2024. P. 6. https://doi.org/10.48550/arXiv.2401.05215.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Naveed H., Khan A.U., Qui S. et al. A comprehensive overview of large language models // arXiv preprint arXiv:2307.06435. 2023. P. 47. https://doi.org/10.48550/arXiv.2307.06435.</mixed-citation><mixed-citation xml:lang="en">Naveed H., Khan A.U., Qui S. et al. A comprehensive overview of large language models. arXiv preprint arXiv:2307.06435. 2023. P. 47. https://doi.org/10.48550/arXiv.2307.06435.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Padmanayana V., Bhavya K. Stock market prediction using Twitter sentiment analysis // Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol. 2021. Vol. 7. No. 4. P. 265–270. https://doi.org/10.32628/CSEIT217475.</mixed-citation><mixed-citation xml:lang="en">Padmanayana V., Bhavya K. Stock market prediction using Twitter sentiment analysis // Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol. 2021. Vol. 7. No. 4. P. 265–270. https://doi.org/10.32628/CSEIT217475.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Schmanski B., Scotti C., Vega C., Benamar H. Fed Communication, News, Twitter, and Echo Chambers. 2023. P. 61. https://doi.org/10.17016/FEDS.2023.036.</mixed-citation><mixed-citation xml:lang="en">Schmanski B., Scotti C., Vega C., Benamar H. Fed Communication, News, Twitter, and Echo Chambers. 2023. P. 61. https://doi.org/10.17016/FEDS.2023.036.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Wei J., Bommasani R., Raffel C. et al. Emergent abilities of large language models // arXiv preprint arXiv:2206.07682. 2022. P. 30. https://doi.org/arXiv.2206.07682.</mixed-citation><mixed-citation xml:lang="en">Wei J., Bommasani R., Raffel C. et al. Emergent abilities of large language models. arXiv preprint arXiv:2206.07682. 2022. P. 30. https://doi.org/arXiv.2206.07682.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Xu G., Meng Y., Qiu X. et al. Sentiment analysis of comment texts based on BiLSTM // Ieee Access. 2019. Vol. 7. P. 51522–51532. https://doi.org/10.1109/ACCESS.2019.2909919.</mixed-citation><mixed-citation xml:lang="en">Xu G., Meng Y., Qiu X. et al. Sentiment analysis of comment texts based on BiLSTM. Ieee Access. 2019. Vol. 7. P. 51522–51532. https://doi.org/10.1109/ACCESS.2019.2909919.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Yasir M., Afzal S., Latif K. et al. An efficient deep learning based model to predict interest rate using twitter sentiment // Sustainability. 2020. Vol. 12. No. 4. P. 1660–1677. https://doi.org/10.3390/su12041660.</mixed-citation><mixed-citation xml:lang="en">Yasir M., Afzal S., Latif K. et al. An efficient deep learning based model to predict interest rate using twitter sentiment. Sustainability. 2020. Vol. 12. No. 4. P. 1660–1677. https://doi.org/10.3390/su12041660.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Githunb. Код, который использовался для написания статьи: URL: https://github.com/BorisenkoGeorgy/Pulse_key_rate_forecast (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Githunb. Kod, kotoryj ispol'zovalsja dlja napisanija stat'i: Available at: https://github.com/BorisenkoGeorgy/Pulse_key_rate_forecast (accessed: 13.05.2025).</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Раздел официального сайта ЦБ РФ, где выложена информация про ключевую ставку ЦБ: URL: https://www.cbr.ru/hd_base/keyrate/ (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Razdel oficial'nogo sajta CB RF, gde vylozhena informacija pro kljuchevuju stavku CB: Available at: https://www.cbr.ru/hd_base/keyrate/ (accessed: 13.05.2025). (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Крупнейший Telegram канал с оперативными новостями: URL: https://t.me/markettwits (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Krupnejshij Telegram kanal s operativnymi novostjami: Available at: https://t.me/markettwits (accessed: 13.05.2025). (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Новость с официального сайта РБК про прогноз ставки ЦБ РФ аналитиками Сбербанка: URL: https://www.rbc.ru/finances/22/10/2019/5daee9689a7947bb17344689 (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Novost' s oficial'nogo sajta RBK pro prognoz stavki CB RF analitikami Sberbanka: Available at: https://www.rbc.ru/finances/22/10/2019/5daee9689a7947bb17344689 (accessed: 13.05.2025). (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Описание модели YandexGPT5: URL: https://habr.com/ru/companies/yandex/articles/885218/ (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Opisanie modeli YandexGPT5: Available at: https://habr.com/ru/companies/yandex/articles/885218/ (accessed: 13.05.2025). (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Информация о самой большой языковой модели с официального сайта Meta: URL: https://ai.meta.com/blog/llama-4-multimodal-intelligence/ (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Informacija o samoj bol'shoj jazykovoj modeli s oficial'nogo sajta Meta: Available at: https://ai.meta.com/blog/llama-4-multimodal-intelligence/ (accessed: 13.05.2025).</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">Карточка модели YandexGPT5 Lite Instruct: URL: https://huggingface.co/yandex/YandexGPT-5-Lite-8B-instruct-GGUF (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Kartochka modeli YandexGPT5 Lite Instruct: Available at: https://huggingface.co/yandex/YandexGPT-5-Lite-8B-instruct-GGUF (accessed: 13.05.2025). (In Russ.).</mixed-citation></citation-alternatives></ref><ref id="cit22"><label>22</label><citation-alternatives><mixed-citation xml:lang="ru">Официальный сайт kaggle: URL: https://www.kaggle.com/ (дата обращения: 13.05.2025).</mixed-citation><mixed-citation xml:lang="en">Oficial'nyj sajt kaggle: Available at: https://www.kaggle.com/ (accessed: 13.05.2025)</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
