مقایسه دقت هوشمندی الگوریتم های مبتنی بر داده کاوی جهت برآورد قیمت ( ارزش) سهام

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی دکتری مدیریت مالی، واحد بین الملل کیش، دانشگاه آزاد اسلامی، جزیره کیش، ایران

2 استادیار گروه مدیریت مالی، واحد بین الملل داشنگاه آزاد کیش، دانشگاه آزاد اسلامی، جزیره کیش، ایران

3 دانشیار گروه مدیریت مالی، واحد بین الملل داشنگاه آزاد کیش، دانشگاه آزاد اسلامی، جزیره کیش، ایران

10.22051/jfm.2024.40333.2685

چکیده

حجم اطلاعات بازار سرمایه به طرز چشمگیری در حال گسترش می‌باشد و بدون استفاده از الگوریتم‌های داده‌کاوی و مدلهای کلان داده، بهره‌برداری از این داده‌ها امکان‌پذیر نخواهد بود. مطالعات گذشته بیانگر امکان پیش‌بینی قیمت سهام توسط مدل‌های یادگیری ماشین می‌باشد؛ اما دقت پیش‌بینی این مدل‌ها مورد ارزیابی قرار نگرفته است. هدف از این پژوهش مقایسه دقت هوشمندی پنج الگوریتم پرکاربرد داده‌کاوی شامل شبکه عصبی، رگرسیون لجستیک، نزدیکترین همسایه k، ماشین بردار پشتیبان و اعتبارسنجی ضربدری می‌باشد. از بین 385 شرکت فعال در بورس اوراق بهادار تهران، 72 شرکت به روش حذف سیستماتیک انتخاب و دقت مدل‌های فوق برای پیش‌بینی قیمت سهام بر روی داده‌های روزانه سهام منتحب برای سال‌های 1388 تا 1399 پیاده‌سازی شده است. متغیر قیمت سهام به عنوان متغیر وابسته و متغیرهای قیمت باز شدن، قیمت بسته شدن، بالاترین قیمت، پایین‌ترین قیمت و حجم معاملات، قیمت روزانه ارز آزاد، قیمت طلا و قیمت نفت به عنوان متغیر مستقل استفاده شده است. برای ارزیابی دقت برآورد قیمت سهام از سه شاخص ، MSE و RMSE استفاده شده و از تحلیل واریانس با استفاده از آماره F برای برازش دقت مدل‌ها و از آماره t برای مقایسه دو به دو مدل‌ها با یکدیگر استفاده شده است. نتایج پژوهش نشان داد از بین الگوریتم‌های هوشمند استفاده شده، الگوریتم ماشین بردار پشتیبان بیشترین قدرت برآورد قیمت سهام را به خود اختصاص داده است

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Comparison of Intelligence Accuracy of Data Mining Algorithms to Estimate Stocks Prices

نویسندگان [English]

  • Hossein Kianizadeh 1
  • Ali Baghani 2
  • Mohsen Hamidian 3
1 Department of Financial Management, Kish International Branch, Islamic Azad University, Kish Island. Iran
2 Department of Financial Management, Kish International Branch, Islamic Azad University, Kish Island. Iran
3 Department of Financial Management, Kish International Branch, Islamic Azad University, Kish Island. Iran
چکیده [English]

Forecasting the stock price due to its attractiveness has always been on the focus of experts and capital market activists. In such a way that various prediction models such as technical and fundamental analysis and data mining are increasingly used to predict stock prices.
Past studies indicate the possibility of stock price prediction by machine learning models, but the prediction accuracy of these models has not been evaluated. The purpose of this research is to accurately compare the intelligence of five commonly used data mining algorithms, including neural network, Logestic regression, k-nearest neighbors, support vector machines and cross validation. Among the 385 active companies in Tehran Stock Exchange, 72 companies have been selected by the method of systematic elimination and the above models have been implemented to predict stock prices on the daily prices of selected stocks for the years 2009 to 2020.
The stock price is used as a dependent variable and changes in the opening price, closing price, highest price, lowest price and volume of trade, daily price of forign currency, gold and oil price are used as independent variables.
Three indicators R2, MSE and RMSE are used, to evaluate the accuracy of models, and analysis of variance using F statistics is used to fit the accuracy of the models, and t-student statistic is used to compare two models.
The results are showed that among the smart algorithms used, the support vector machine algorithm has the highest accuracy. Matlab software is used to analyze the data and compare the models.

کلیدواژه‌ها [English]

  • Stock exchange
  • Intelligent algorithms
  • Machine learning
  • Data mining
 
Ahmadkhan Beigi, Sohail, Abdulvand, Neda. (2016). Stock price forecasting with a combined approach of artificial neural networks and colonial competition algorithm based on chaos theory. Financial Management Strategy, 5 (3), 27- 73. Doi: 10. 22051/jfm. 2017. 14635. 1319. (In Persian).
Afshari Rad, Elham, Alavi, Seyed Enayat A.and Sinaii, Hassan Ali. (2017). "An intelligent model for predicting stock trends using technical analysis methods" Financial Research, Summer 2017, Volume 20, Number 249. (in persian)
Ahmadkhan Beigi, Sohail and Abdulvand, Neda. (2016). "Stock price forecasting with a hybrid approach of artificial neural networks and colonial competition algorithm based on chaos theory." Financial management strategy, number 5. (in persian)
Ampomah, E. K., Z. Qin, and G. Nyame. (2020). "Evaluation of tree-based ensemble machine learning models in predicting stock price direction of movement‏." Information, 11(6), 332.
Ballings, M, D Van den Poel, N Hespeels, and R. Gryp. (2015). "Evaluating multiple classifiers for stock price direction prediction.‏" Expert systems with Applications, 42(20), 7046-7056.
Bonaccorso, G. (2017). Machine learning algorithms‏. Packt Publishing Ltd.
Ebrahimi Heravi, Behrouz, Kazem Rangzan, Mostafa Kabuli, Hassan Daneshian. (2016). "Comparison of Artificial Neural Network and Fuzzy System Methods in Determining Flood Early Warning Time of the Yellow River Watershed Sub-basin - Khuzestan Province" Geography and Environmental Planning, Volume 28, Number 1, Pages 1-20. (in persian)
Emami, Mohammad, Yatharbi, Seyyed Shahabuddin. (2013). "The use of artificial neural network in the interpretation of barometric test results". Scientific Research Journal of Imran Modares, 14th period, special issue (in persian)
Gardening, Shahnaz. (2016). "Techniques and methods of machine learning on big data." National conference of new technologies in electrical and computer engineering. (in persian)
Hafezi, R., J. Shahrabi, and E. Hadavandi. (2015). "A bat-neural network multi-agent system (BNNMAS) for stock price prediction: Case study of DAX stock price. ‏." Applied Soft Computing, 29 196-210.
Hosseininasab, Hojjat, Karimi Teklo, Salim and Yousefinejad, Marzieh. (2012). "Comparison of the accuracy of support vector machines and artificial neural networks approaches in predicting the profit per share of companies listed on the Tehran Stock Exchange." Bi-Quarterly Economic Essays, 109-134-10 (in persian)
Houssein, E. H., M. Dirar, K. Hussain, and W. M. mohamed. (2021). "Artificial Neural Networks for Stock Market Prediction: A Comprehensive Review‏." Metaheuristics in Machine Learning: Theory and Applications 409-444.
Kantardzik, Mehmed. (2012). Data mining, translated by Amir Alikhanzadeh, Computer Sciences Publishing. (in persian)
Lin, Q. (2018). "Technical analysis and stock return predictability: An aligned approach‏." Journal of financial markets, 38 103-123.
Long, W., L Song, and Y Tian. (2019). "A new graphic kernel method of stock price trend prediction based on financial news semantic and structural similarity‏." Expert Systems with Applications, 118 411-424.
Nivetha, R. Y., and C. Dhaya. (2017). "Developing a prediction model for stock analysis.‏" International Conference on Technical Advancements in Computers and Communications (ICTACC). IEEE.
Piryonesi, S. M., and T. E. El-Diraby. (2020). "Data analytics in asset management: Cost-effective prediction of the pavement condition index‏." Journal of Infrastructure Systems, 26(1), 04019036.
Rai, Reza, Pouyanfar, Ahmed. (1400), Advanced Investment Management. Side Publications (in persian)
Sarmad, Zahra, Bazargan, Abbas and Hijazi, Elaha. (2019). Research methods in behavioral sciences. Post an ad. (in persian)
Sigo, M. O., M. Selvam, S. Venkateswar, and C. Kathiravan. (2020). "Application of ensemble machine learning in the predictive data analytics of indian stock market." CIFR Paper Forthcoming.‏
Wen Long Zhichen Lu Lingxiao Cui. (2019). "Deep learning-based feature engineering for stock price movement prediction،." Knowledge-Based Systems Volume 164، 163-173.
Zarei, Qasim, Mohammadian, Rana, Hashedi Neiri, Hatef and Ajirlu, Mohammad. (2017) "Comparison of fuzzy neural network methods with fuzzy wavelet neural networks in predicting stock prices of banks listed on the Tehran Stock Exchange" Journal of Financial Management Strategy, Fall 6th Year 2017 Number 2. (in persian)
Zhou, X, L Wang, H Liao, S Wang, and B Lev. (2019). "A prospect theory-based group decision approach considering consensus for portfolio selection with hesitant fuzzy information