مدیریت بازاریابی هوشمند

مدیریت بازاریابی هوشمند

مدل هوشمند پیش‌بینی قیمت سهام مبتنی بر یادگیری عمیق: با رویکرد کاهش ابعاد

نوع مقاله : استخراج از رساله دکتری

نویسندگان
1 دانشجوی دکتری مدیریت فناوری اطلاعات، گروه مدیریت فناوری اطلاعات، دانشکده مدیریت وحسابداری، واحد قزوین، دانشگاه آزاد اسلامی،
2 استادیار گروه مدیریت فناورری اطلاعات، دانشکدۀ مدیریت و حسابداری، واحد قزوین، دانشگاه آزاد اسلامی، قزوین، ایران
3 استادیار گروه اقتصاد، دانشکده مدیریت وحسابداری، واحد قزوین، دانشگاه آزاد اسلامی، قزوین، ایران
چکیده
پیش‌بینی قیمت و بازده سهام یکی از پیچیده‌ترین و بحث‌برانگیزترین موضوعات در بازارهای مالی است. بازار سهام همواره تحت تأثیر وضعیت اقتصاد ملی، تصورات سرمایه‌گذاران و رویدادهای سیاسی بوده است و سری قیمت‌ها به شدت غیرخطی و غیرثابت است. با تحقیقات مستمر و به‌روزرسانی محققان در بازار اقتصادی و تئوری بازار سهام، به‌تدریج مؤلفه‌های پیش‌بینی شاخص قیمت سهام در معرض دید قرار گرفتند و پیش‌بینی قیمت سهام امکان‌پذیر شد. این پژوهش نیز با هدف ارائۀ مدل هوشمند پیش‌بینی قیمت سهام مبتنی بر یادگیری عمیق در بازار بورس اوراق بهادار تهران ـ با رویکرد تکنیک‌های کاهش ابعاد جهت مدیریت سبد سرمایه انجام شد تا افزایش بازده و کاهش ریسک سرمایه‌گذاری ممکن شود. داده‌های استفاده‌شده در بازۀ زمانی 1399 تا 1402 از سامانۀ کدال دریافت و با استفاده از روش CRISP و با استفاده از زبان برنامه‌نویسی پایتون کدنویسی و تحلیل شدند. برای مدل پیشنهادی ترکیبی از الگوریتم‌های LSTM، PCA و SVD استفاده شد. مقایسۀ ترکیب روش‏های کاهش ابعاد با روش‏های هوش مصنوعی نشان می‏دهد برای پیش‌بینی برای روز آتی به‌کارگیری روش کاهش ابعاد PCA می‏تواند عملکرد یادگیری عمیق را نسبت‌به سایر روش‏های کاهش ابعاد داده بهبود بخشد.
کلیدواژه‌ها

موضوعات


عنوان مقاله English

An intelligent stock price forecasting model based on deep learning: with dimensionality reduction approach

نویسندگان English

Shohreh Zokaei 1
MohammadReza Sanaei 2
Akbar Mirzapour Babajan 3
1 PhD Candidate of Information Technology Management, Department of Information Technology Management, Faculty of Management and Accounting, Qazvin Branch, Islamic Azad University, Qazvin, Iran
2 Assistant Professor, Department of economy, Faculty of Management and Accounting, Qazvin Branch, Islamic Azad University, Qazvin, Iran
3 Assistant professor,Department Economics, Faculty of Management and Accounting, Qazvin Branch, Islamic Azad University, Qazvin, Iran
چکیده English

Forecasting stock price and returns is one of the most complicated andcontroversial issues in financial markets . The stock market has always been influenced by the state of the national economy, investors; perceptions and political events, and the price series is highly non-linear and unstable. With continuous research and updating of researchers in the economic market and stock market theory, the components of stock price index prediction were gradually exposed and stock price prediction became possible. This research was also conducted with the aim of providing an intelligent stock price forecasting model based on deep learning in the Tehran Stock Exchange market - with the approach of dimensionality reduction techniques for managing the capital portfolio in order to increase returns and reduce investment risk . The data used in the period of 2020-2023 were received from the Kodal system and coded and analyzed using the CRISP method and using the Python programming language. A combination of LSTM, PCA and SVD algorithms was used for the proposed model. Comparing the combination of dimensionality reduction methods with artificial intelligence methods shows that the use of PCA dimensionality reduction method can improve the performance of deep learning compared to other data dimensionality reduction methods .

کلیدواژه‌ها English

Intelligent model
stock price prediction
deep learning
 شریف‌فر، امیر، خلیلی، مریم، رئیسی وانانی، ایمان، میرفیض فلاح، شمس (1400). کاربرد معماری‌های یادگیری عمیق در پیش‌بینی قیمت سهام (رویکرد شبکۀ عصبی پیچشی)؛ مدیریت دارایی و تامین مالی، 38، 1ـ20.
Bauckhage, C., Kersting, K., Hadiji, F. (2013). Mathematical models of fads explain the temporal dynamics of internet memes, in Seventh International AAAI Conference on Weblogs and Social Media.
Beg¨um, D., Lorenzo, B.  (2015). A Novel Active Learning Method in Relevance Feedback for Contentbased Remote Sensing Image Retrieval. IEEE Transactions on Geoscience and Remote Sensing, 53, 5, 2323–2334..
Christoph, K., Rodner, E., Freytag, A., Denzler, J. (2016). Active and Continuous Exploration with Deep Neural Networks and Expected Model Output Changes. arXiv preprint arXiv:1612.06129.
Chung, H., Shin, K. S. (2018). Genetic algorithm ـ optimized long short ـ term memory network for stock market prediction. Sustain, vol. 10, no. 10.
Chung, H., Shin, K. S. (2020). Genetic algorithm ـ optimized multi ـ channel convolutional neural network for stock market prediction. Neural Comput. Appl., vol. 32, no. 12, 7897–7914.
Guoliang, H., Li, Y., Zhao, W. (2017). An Uncertainty and Density based Active Semi ـ supervised Learning Scheme for Positive Unlabeled Multivariate Time Series Classification. Knowledge ـ Based Systems, 124 (2017), 80–92..
Jiang, W. (2020). Applications of deep learning in stock market prediction: recent progress, Elsevier J., pp. 1–97.
Kaiming, H., Zhang, X., Ren, Sh., Sun., J. (2016). Deep Residual Learning for Image Recognition.. In International Conference on Computer Vision and Pattern Recognition. 770–778.
Kanwal, A., et al. (2023). BiCuDNNLSTM-1dCNN—A hybrid deep learning-based predictive model for stock price prediction, Expert Systems With Applications, vol.202, pp 17-31.
Li, X., Xie, H., Lau, R. Y. K., Wong, T. L., Wang, F. L. (2018). Stock Prediction via Sentimental Transfer Learning. IEEE Access, vol. 6, 73110–73118.
Lilly, Sh., Gupta, S., Ragavender, N., Anirudh, R. M., Divya, D., (2021). Time Series Model for Stock Market Prediction Utilising Prophet. Turkish Journal of Computer and Mathematics Education; Trabzon , 12(6), 4529 ـ 4534.
Liu, H., Long, Z. (2020). An improved deep learning model for predicting stock market price time series. Digit. Signal Process. A Rev. J., vol. 102, p. 102741.
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., et al., (2016). SSD: Single shot multibox detector.. Lecture notes in computer science (including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics).
Long, W., Lu, Z., Cui, L. (2019). Deep learning ـ based feature engineering for stock price movement prediction. Knowledge ـ Based Syst., vol. 164, 163–173.
Michiel, H. (2013). Encyclopaedia of Mathematics:. Volume 6: Subject IndexAuthor Index. Springer Science & Business Media..
Pang, X., Zhou, Y., Wang, P., Lin, W., Chang, V. (2020). An innovative neural network approach for stock market prediction. Supercomput., vol. 76, no. 3, 2098–2118.
Paulina, H., Masko, D. (2015). The Impact of Imbalanced Training Data for Convolutional Neural Networks.. Degree Project in Computer Science, KTH Royal Institute of Technology .
Sinaga, F. M., Felix, M. J., Halim, A. (2019). Stock Trend Prediction using SV ـ kNNC and SOM. Proc. 2019 4th Int. Conf. Informatics Comput. ICIC 2019.
Singh, S., Ahmad, M., Bhattacharya, A., Azhagiri, M. (2019). Predicting stock market trends using hybrid SVM model and LSTM with sentiment determination using natural language processing. Int. J. Eng. Adv. Technol., vol. 9, no. 1, 2870–2875.
Xu, Y., Yang, C., Peng, S., Nojima, Y. (2020). A hybrid two ـ stage financial stock forecasting algorithm based on clustering and ensemble learning. Appl. Intell.
Yarin, G., Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In International Conference on Machine Learning. 1050–1059.
Zahedi, M. S., Bokaei, M. H., Shoeleh, F., Yadollahi, M. M., Doostmohammadi, E., Farhoodi, M. )2018(. Persian word embedding evaluation benchmarks. Electrical Engineering (ICEE), Iranian Conference, pp. 1583–1588,.
Zhang, G. P.  (2007). A neural network ensemble method with jittered training data for time series forecasting. Inf. Sci., 177(23), 5329–5346.
Zhang, J., Teng, Y. F., Chen, W. (2019). Support vector regression with modified firefly algorithm for stock price forecasting. Appl. Intell., vol. 49, no. 5, 1658–1674.
Zhang, R., Yuan, Z., Shao, X. (2018). A New Combined CNN ـ RNN Model for Sector Stock Price Analysis. ” Proc.  ـ  Int. Comput. Softw. Appl. Conf., vol. 2, 546–551 .
Zhang, Y., Yan, B., Aasma, M. (2020). A novel deep learning framework: Prediction and analysis of financial time series using CEEMD and LSTM. Expert Syst. Appl., vol. 159, 113609.

  • تاریخ دریافت 10 اردیبهشت 1403
  • تاریخ بازنگری 17 شهریور 1403
  • تاریخ پذیرش 21 مهر 1403