<?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">uzria</journal-id><journal-title-group><journal-title xml:lang="ru">Ученые записки Российской академии предпринимательства</journal-title><trans-title-group xml:lang="en"><trans-title>Scientific notes of the Russian academy of entrepreneurship</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">2073-6258</issn><publisher><publisher-name>JSC “Publishing Agency “Science and Education”</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.24182/2073-6258-2021-20-3-52-63</article-id><article-id custom-type="elpub" pub-id-type="custom">uzria-686</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>ЭКОНОМИЧЕСКИЙ РОСТ: ПРОБЛЕМЫ И ПЕРСПЕКТИВЫ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>ECONOMIC GROWTH: PROBLEMS AND PROSPECTS</subject></subj-group></article-categories><title-group><article-title>Сравнительный анализ качества прогнозирования классической статистической модели и модели машинного обучения на данных российского фондового рынка</article-title><trans-title-group xml:lang="en"><trans-title>Comparative analysis of the forecasting quality of the classical statistical model and the machine learning model on the data of the Russian stock market</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>Щербинина</surname><given-names>А. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Shcherbinina</surname><given-names>A. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Специалист</p><p>Москва</p></bio><bio xml:lang="en"><p>Specialist</p><p>Moscow</p></bio><email xlink:type="simple">scherbinina.av@rea.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Алжеев</surname><given-names>А. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Alzheev</surname><given-names>A. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Магистр</p><p>Москва</p></bio><bio xml:lang="en"><p>Masterstudent</p><p>Moscow</p></bio><email xlink:type="simple">alzheev@gmail.com</email><xref ref-type="aff" rid="aff-2"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Кафедра маркетинга, Российский экономический университет В.Г. Плеханова</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Department of Marketing, Plekhanov Russian University of Economics</institution><country>Russian Federation</country></aff></aff-alternatives><aff-alternatives id="aff-2"><aff xml:lang="ru"><institution>Факультет прикладной математики и информационных технологий, Финансовый университет при Правительстве Российской Федерации</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Faculty of Applied Mathematics and Information Technology, Financial University under the Government of the Russian Federation</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2021</year></pub-date><pub-date pub-type="epub"><day>05</day><month>10</month><year>2021</year></pub-date><volume>20</volume><issue>3</issue><fpage>52</fpage><lpage>63</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Щербинина А.В., Алжеев А.В., 2021</copyright-statement><copyright-year>2021</copyright-year><copyright-holder xml:lang="ru">Щербинина А.В., Алжеев А.В.</copyright-holder><copyright-holder xml:lang="en">Shcherbinina A.V., Alzheev A.V.</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://www.scinotes.ru/jour/article/view/686">https://www.scinotes.ru/jour/article/view/686</self-uri><abstract><p>Основная задача данной работы — сравнение прогностической способности классической модели машинного обучения — ARIMA, как наиболее распространенной и хорошо изученной baseline модели, и ML модели на основе последовательной нейронной сети — в данном случае LSTM. Целью является максимизация точности и минимизация ошибки — подбор наиболее подходящей модели для прогнозирования временных рядов с наивысшей точностью. Для данных математических моделей приведено описание. Также предложен алгоритм для прогноза временных рядов в рамках использования данных моделей, основанный на подходе «Rolling window» («скользящее окно»). Практическая имплементация реализована с использованием среды программирования Python с библиотеками Pandas, Numpy, pmdarima, Keras, Statsmodels. Для обучения моделей использованы биржевые данные по цене закрытия за акцию ведущих российских компания: Яндекс, ВТБ, КамАЗ, Киви, Газпром, НЛМК, Роснефть, Алроса. Проведённые исследования демонстрируют прогностическое превосходство подхода, основанного на нейронных сетях, при этом среднеквадратическая ошибка RMSE на 71% меньше аналогичного показателя для модели ARIMA, что позволяет сделать вывод о предпочтительности использования модели LSTM для данного класса задач.</p></abstract><trans-abstract xml:lang="en"><p>The main objective of this work is to compare the predictive ability of the classical machine learning model — ARIMA, as the most common and well-studied baseline model, and the ML model based on a sequential neural network — in this case, LSTM. The goal is to maximize accuracy and minimize error — selecting the most appropriate model for predicting time series with the highest accuracy. A description is given for these mathematical models. An algorithm is also proposed for forecasting time series using these models, based on the «Rolling window» approach. Practical implementation is implemented using the Python programming environment with the Pandas, Numpy, pmdarima, Keras, Statsmodels libraries. To train the models, we used stock data at the closing price per share of the leading Russian companies: Yandex, VTB, KamAZ, Kiwi, Gazprom, NLMK, Rosneft, Alrosa for the period. The studies carried out demonstrate the predictive superiority of the approach based on neural networks, while the RMSE is 71% less than the same indicator for the ARIMA model, which allows us to conclude that the use of the LSTM model is preferable for this class of problems.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>фондовые рынки</kwd><kwd>модель машинного обучения</kwd><kwd>эконометрика</kwd><kwd>ARIMA</kwd><kwd>LSTM</kwd><kwd>алгоритм</kwd></kwd-group><kwd-group xml:lang="en"><kwd>stock markets</kwd><kwd>machine learning model</kwd><kwd>econometrics</kwd><kwd>ARIMA</kwd><kwd>LSTM</kwd><kwd>algorithm</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">Магнус Я.Р., Катышев П.К., Пересецкий А.А. Эконометрика. Начальный курс. – М.: Дело; 2007. 504 с.</mixed-citation><mixed-citation xml:lang="en">Magnus Ya.R., Katyshev P.K., Peresetskii A.A. Ekonometrika. Nachal’nyi kurs. – M.: Delo; 2007. 504 s.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Schmidhuber J. Habilitation thesis: System modeling and optimization. – Munich: Technical University of Munich; 1993. 209 p. (In Germ.)</mixed-citation><mixed-citation xml:lang="en">Schmidhuber J. Habilitation thesis: System modeling and optimization. – Munich: Technical University of Munich; 1993. 209 p. (In Germ.).</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Hochreiter S., Schmidhuber J. Long short-term memory. Neural Computation. 1997; 9(8): 1735–1780. DOI:10.1162/neco.1997.9.8.1735. PMID 9377276.</mixed-citation><mixed-citation xml:lang="en">Hochreiter S., Schmidhuber J. Long short-term memory. Neural Computation. 1997; 9(8): 1735–1780. DOI:10.1162/neco.1997.9.8.1735. PMID 9377276.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Brownlee J. Time Series Prediction with LSTM Recurrent Neural Networks in Python with Keras. Machine Learning Mastery. 2016; https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras.</mixed-citation><mixed-citation xml:lang="en">Brownlee J. Time Series Prediction with LSTM Recurrent Neural Networks in Python with Keras. Machine Learning Mastery. 2016; https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Gers F.A., Schmidhuber J., Cummins F. Learning to Forget: Continual Prediction with LSTM. Neural Computation. 2000; 12(10): 2451–2471.</mixed-citation><mixed-citation xml:lang="en">Gers F.A., Schmidhuber J., Cummins F. Learning to Forget: Continual Prediction with LSTM. Neural Computation. 2000; 12(10): 2451–2471.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Кратович П.В. Нейронные сети и модели ARIMA для прогнозирования котировок. Программные продукты и системы. 2011; (1): 95–98.</mixed-citation><mixed-citation xml:lang="en">Kratovich P.V. Neironnye seti i modeli ARIMA dlya prognozirovaniya kotirovok. Programmnye produkty i sistemy. 2011; (1): 95–98.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Garcia1 F., Guijarro F., Moya I., Oliver J. Estimating returns and conditional volatility: a comparison between the ARMA-GARCH-M Models and the Backpropagation Neural Network. Int. J. Complex Systems in Science. 2012; 1(2): 21–26. ISSN 2174-6036.</mixed-citation><mixed-citation xml:lang="en">Garcia1 F., Guijarro F., Moya I., Oliver J. Estimating returns and conditional volatility: a comparison between the ARMA-GARCH-M Models and the Backpropagation Neural Network. Int. J. Complex Systems in Science. 2012;1(2):21-26. ISSN 2174-6036.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Krauss C., Do X.A., Huck N. Deep neural networks, gradientboosted trees, random forests: Statistical arbitrage on the S&amp;P 500. European Journal of Operational Research. 2016; 259(2). DOI:10.1016/j.ejor.2016.10.031.</mixed-citation><mixed-citation xml:lang="en">Krauss C., Do X.A., Huck N. Deep neural networks, gradientboosted trees, random forests: Statistical arbitrage on the S&amp;P 500. European Journal of Operational Research. 2016; 259(2). DOI:10.1016/j.ejor.2016.10.031.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Chung J., Lee D., Seo Y., Yoo C.D. Deep attribute networks. NIPS Workshop on Deep Learning and Unsupervised Feature Learning. 2012; 3(2012). http://www.eng.uwaterloo.ca/~jbergstr/files/nips_dl_2012/Paper%2011.pdf.</mixed-citation><mixed-citation xml:lang="en">Chung J., Lee D., Seo Y., Yoo C.D. Deep attribute networks. NIPS Workshop on Deep Learning and Unsupervised Feature Learning. 2012; 3(2012). http://www.eng.uwaterloo.ca/~jbergstr/files/nips_dl_2012/Paper%2011.pdf.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Fischera T., Krauss C. Deep Learning with Long Short-term Memory Networks for Financial Market Predictions. Eur. J. Oper. Res. 2018; 270(2): 654–669.</mixed-citation><mixed-citation xml:lang="en">Fischera T., Krauss C. Deep Learning with Long Short-term Memory Networks for Financial Market Predictions. Eur. J. Oper. Res. 2018; 270(2): 654–669.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Эконометрика. Под ред. Елисеевой И.И. – М.: Финансы и статистика; 2006. 576 с. ISBN 5-279-02786-3.</mixed-citation><mixed-citation xml:lang="en">Ekonometrika. Pod red. Eliseevoi I.I. – M.: Finansy i statistika, 2006. 576 s. ISBN 5-279-02786-3.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Armstrong J.S., Collopy F. Error Measures For Generalizing About Forecasting Methods: Empirical Comparisons. International Journal of Forecasting. 1992; 8(1): 69–80. DOI:10.1016/0169-2070(92)90008-w.</mixed-citation><mixed-citation xml:lang="en">Armstrong J.S., Collopy F. Error Measures For Generalizing About Forecasting Methods: Empirical Comparisons. International Journal of Forecasting. 1992; 8(1): 69–80. DOI:10.1016/0169-2070(92)90008-w.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Hyndman R.J., Athanasopoulos G. Forecasting: Principles and Practice. 2 edition. Melbourne, Australia: OTexts; 2018. 382 p. ISBN-13: 978-0987507112.</mixed-citation><mixed-citation xml:lang="en">Hyndman R.J., Athanasopoulos G. Forecasting: Principles and Practice. 2 edition. Melbourne, Australia: OTexts; 2018. 382 p. ISBN-13: 978-0987507112.</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>
