الگوریتم پیشنهادی برای انتخاب ویژگی‌های مناسب به‌منظور پیش‌بینی شاخص بورس اوراق بهادار تهران

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی دکتری مدیریت مالی، دانشگاه شهید یهشتی، تهران، ایران.

2 دانشیار، گروه مدیریت مالی و بیمه، دانشگاه شهید یهشتی، تهران، ایران

3 استاد، گروه مدیریت بازرگانی، دانشگاه شهید بهشتی، تهران،ایران.

چکیده

عملکرد یک مدل هوشمند تا حد زیادی به انتخاب مرتبط­ترین و تأثیرگذارترین متغیرهای ورودی و کمترین پیچیدگی مدل یادگیری بستگی دارد. از این­رو در مطالعه حاضر، برای پیش­بینی روزانه شاخص کل بورس اوراق بهادار تهران بر اساس متغیرهای مالی و اقتصادی، ابتدا اقدام به اولویت­بندی ویژگی­ها با الگوریتم MID نموده، سپس از 4 مدل مختلف شبکه عصبی (MLP, SVR, RBF, DNN) که از مهم­ترین و بدیع­ترین مدل­های پیش­بینی می­باشند، استفاده می­شود. با توجه به نتایج بدست آمده از تحلیل مدل­های مورد بررسی، در نهایت الگوریتمی برای انتخاب ویژگی­های مناسب برای پیش­بینی شاخص، تحت عنوانISF­_MID پیشنهاد شده و با تعدادی از روش­های مشابه، مقایسه می­گردد. داده­های مورد استفاده در این پژوهش به صورت روزانه در بازه زمانی 28/10/1392 تا 30/5/1397 جمع­آوری شده­اند. مدل­های مورد بررسی در مرحله پیاده­سازی با روش اعتبارسنجی متقابل K-fold مورد ارزیابی قرار گرفتند. همچنین از معیارهای MAE، MSE و RMSE برای ازریابی عملکرد مدل­های مذکور استفاده می­شود. نتایج نشان می­دهد که با روش­ پیشنهادی، می­توان با 7 ویژگی انتخابی به دقت بالایی در پیش­بینی روزانه شاخص کل بورس اوراق بهادار تهران دست­یافت.

کلیدواژه‌ها


عنوان مقاله [English]

The Proposed Algorithm to Select Appropriate Features for Predicting Tehran Stock Exchange Index

نویسندگان [English]

  • Somayeh Mohebbi 1
  • Mohamad Esmaeil Fadaeinejad 2
  • Mohammad reza Hamidizadeh 3
1 Ph.D. Candidate in Financial Management, Shahid Beheshti University, Tehran, Iran.
2 Associate Prof., Department of Financial Management and Insurance, Shahid Beheshti University, Tehran, Iran.
3 Professor, Department of Business Management, Shahid Beheshti University, Tehran, Iran.
چکیده [English]

The performance of an intelligent model largely depends on the selection of the most relevant and most influential input variables and the lowest complexity of the learning model. Therefore, in the present study, to predict the index of Tehran Stock Exchange based on financial and economic variables, first prioritize featuresWith MID algorithm, then 4 different neural network models (MLP, SVR, RBF, DNN) are used, which are the most important and innovative prediction models. According to the results of the analysis of the studied models, an algorithm is proposed to select the appropriate features on the index, as ISF-MID, and are compared with several similar methods. The data used in this study were collected daily in the period of 18/01/2014 to 21/08/2018. Evaluation of the models was performed by K-fold cross validation method. The MAE, MSE, and RMSE criteria are also used to evaluate the performanceof the mentionedmodels. The results show that with the proposed method, with 7 selected features, it is possible to achieve high accuracy in predicting the daily index of the Tehran Stock Exchange.

کلیدواژه‌ها [English]

  • : Stock Index Prediction
  • Feature Selection Algorithm
  • Radial Basis Function
  • Support Vector Regression
  • Deep Neural Network
  1. Arévalo, R., García, J., Guijarro, F., & Peris, A. )2017(. A dynamic trading rule based on filtered flag pattern recognition for stock market price forecasting . Expert Systems with Applications, 81, 177-192.
  2. Asadi, Sh., Hadavandi, E., Mehmanpazir, F., & Nakhostin, M. M. (2012). Hybridization of evolutionary Levenberg-Marquardt neural networks and data pre-processing for stock market prediction. Knowledge-Based Systems, 35, 245-258.
  3. Atsalakis, G. S., & Valavanis, K. P. (2009). Surveying stock market forecasting techniques -Part II: Soft computing methods. Expert Systems with Applications, 36(3), 5941-5950.
  4. Badiei, H., Rezazadeh, R., & Mahmoudi, H. (2017). Forecasting Stock Price Trend by Artificial Neural Networks (Case Study: Isfahan Oil Refinery Company). Financial Engineering and Portfolio Management, 8(31), 167-185. (in Persian)
  5. Bajalan, S., Fallahpour, S., & Dana, N. (2016). The Stock Trend Prediction Using Volume Weighted Support Vector Machine with a Hybrid Feature Selection Method to Predict the Stock Price Trend in Tehran Stock Exchange.Financial Management Strategy, 4(3), 121-148. (in Persian)
  6. Berrar, D. (2018). Cross-validation. Encyclopedia of Bioinformatics and Computational Biology, 1, 542-545.
  7. Botshekan, M., & Mohseni, H. (2018). Investigation volatility spillovers between oil market and stock index return.Journal of Investment Knowledge, 7(25), 267-284. (in Persian)
  8. Cavalcante, R. C., Brasileiro, R. C., Souza, V. L. F., Nobrega, J. P., & Oliveira A. L. I. (2016). Computational Intelligence and Financial Markets: A Survey and Future Directions, Expert Systems With Applications, 55, 194-211.
  9. Chong, E., Han, Ch., & Park F. (2017). Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies, Expert Systems With Applications, 83,187-205.
  10. Ding, C., & Peng, H. (2005). Minimum redundancy feature selection from microarray gene expression data. Journal of bioinformatics and computational biology, 3(2), 185-205.
  11. Enke, D., Thawornwong, S. (2005). The use of data mining and neural networks for forecasting stock market returns. Expert Systems with Applications, 29(4), 927-940.
  12. Faghihi Nezhad, M., & Minaei, B. (2018). Prediction of Stock Market Behavior Based on Artificial Neural Networks through Intelligent Ensemble Learning Approach.Industrial Management Journal, 10(2), 315-334. (in Persian)
  13. Fallahpour, S., Golarzi, G., & Fatourechian, N. (2013). Predicting Stock Price Movement Using Support Vector Machine Based on Genetic Algorithm in Tehran Stock Exchange Market. Financial Research Journal, 15(2), 269-288. (in Persian)
  14. Grigoryan, H. (2016). Stock Market Prediction Method Based on Support Vector Machines (SVM) and Independent Component Analysis (ICA). Database Systems Journal, 7(1), 12-21.
  15. Gunduz, H., & Cataltepe, Z. (2015). Borsa Istanbul (BIST) daily prediction using financial news and balanced feature selection. Expert Systems with Applications, 42 (22), 9001-9011.
  16. Guresen, E., Kayakutlu, G., & Daim, T. U. (2011). Using artificial neural network models in stock market index prediction. Expert systems with application, 38, 10389-10397.
  17. Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Literature review: Machine learning techniques applied to financial market prediction. Expert Systems with Applications, 124, 226-251.
  18. Hinton, G. E., & Salakhutdinov, R. R. (2006). Reducing the dimensionality of data with neural networks. Science, 313(5786), 504-507.
  19. Hiransha, M., Gopalakrishnan, E. A., Menon, V. K., & Soman, Kp. (2018). NSE stock market prediction using deep-learning models. Procedia Computer Science, 132: 1351-62.
  20. Hu, Y., Liu, K., Zhang, X., Su, L., Ngai, E. W. T., & Liu, M. (2015). Application of evolutionary computation for rule discovery in stock algorithmic trading: A literature review. Applied Soft Computing, 36, 534-51.
  21. Huang, C. F. (2012). A hybrid stock selection model using genetic algorithms and support vector regression. Applied Soft Computing, 12(2), 807-818.
  22. Huang, C. L., & Tsai, C. Y. (2009). A hybrid SOFM-SVR with a filter-based feature selection for stock market forecasting. Expert Systems with Applications, 36(2), 1529-1539.
  23. Lam, M. (2004). Neural network techniques for financial performance prediction: Integrating fundamental and technical analysis. Decision Support Systems, 37(4), 567-581.
  24. Lee, H., Grosse, R., Ranganath, R., & Ng, A. Y. (2009). Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. In Proceedings of the 26th annual international conference on machine learning, (pp. 609-616). Montreal, Canada.
  25. Lee, M. C. (2009). Using support vector machine with a hybrid feature selection method to the stock trend prediction. Expert Systems with Applications, 36(8),10896-10904.
  26. Long, W., Lu, Z., & Cui, L. (2019). Deep learning-based feature engineering for stock price movement prediction. Knowledge-Based Systems, 164, 163-173.
  27. Mandal, M., & Mukhopadhyay, A. (2013). An improved minimum redundancy maximum relevance approach for feature selection in gene expression data. Procedia Technol,10, 20-27.
  28. Monfared, J., & Alinejad, M., & Metghalchi, S. (2012). A Comparative Study of Neural Network Models With Box Jenkins Methodologies in Prediction of Tehran Price Index(TEPIX). Financial Engineering and Securities management (Portfolio Management), 3(11), 1-16. (in Persian)
  29. Nguyen, T. H., Shirai, K., & Velcin, J. )2015(. Sentiment Analysis on Social Media for Stock Movement Prediction. Expert Systems with Applications, 42(24), 9603-9611.
  30. Ou, P., & Wang, H. (2009). Prediction of stock market index movement by ten data mining techniques. Modern Applied Science, 3(12), 28-42.
  31. Oussar, Y., & Dreyfus, G. (2000). Initialization by selection for wavelet network training. Neurocomputing, 34(1), 131-143.
  32. Pakdin Amiri, A., Pakdin Amiri, M., & Pakdin Amiri, M. (2009). Introducing a Prediction Model in Total Stock Price Index Using Neural Networks Approach (Case Study: Tehran Stock Exchange). Journal of Iran's Economic Essays, 6(11), 83-108. (in Persian)
  33. Pashootanizadeh, H., Ranaei Kordshouli, H., Abbasi, A., Moosavihaghighi, M. (2020). Simulation the Model of Effects of Behavioral and Macroeconomic Factors on the Tehran Stock Exchange Index with Using System Dynamics Approach. ـJournal of Financial Management Perspective, 10(29), 89-124. 
  34. Perez-Rodriguez, J. V., Torrab, S., & Andrada-Felixa, J. (2004). STAR and ANN models: Forecasting performance on the Spanish Ibex-35 stock index. Journal of Empirical Finance, 12(3), 490-509.
  35. Qiu, M., & Song, Y. (2016). Predicting the Direction of Stock Market Index Movement Using an Optimized Artificial Neural Network Model. PLoS ONE, 11(5), 1-11.
  36. Raee, R., Nikahd, A., & Habibi, M. (2017). The Index Prediction of Tehran Stock Exchange by Combining the Principal Components Analysis, Support Vector Regression and Particle Swarm Optimization.Financial Management Strategy, 4(15), 1-23. (in Persian)
  37. Raoofi, A., & Mohammadi, T. (2018). Forecasting Tehran Stock Exchange Index Returns Using a Combination of Wavelet Decomposition and Adaptive Neural Fuzzy Inference Systems. Iranian Journal of Economic Research, 23(76), 107-136. (in Persian)
  38. Salehi, M., & Garshasbi, F. (2019). Tehran Stock Exchange Index Forecasting Using Approach Adaptive Neural-Fuzzy Inference System and Imperialist Competitive Algorithm. IT Management Studies, 8(29), 5-34. (in Persian)
  39. Sezer, O. B., Gudelek, M. U., & Ozbayoglu, A. M. (2020). Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. Applied Soft Computing, 90, 1-63.
  40. Singh, R., & Srivastava, S. (2017). Stock prediction using deep learning. Multimedia Tools and Applications, 76(18),18569-18584.
  41. Sorzano, C. O. S., Vargas, J., & Pascual-Montano, A. (2014). A survey of dimensionality reduction techniques. Cornell University Library Abstracts, 1-35.
  42. Tehrani, R., & Moradpour, S. (2012). Predicting the return of the stock market index using base radius artificial neural network models. Financial Engineering and Securities Management, 3(10), 75-90. (in Persian)
  43. Tehrani, R., Heyrani, M., & Mansuri, S. (2019). A Comparison between Fama and French five-factor model and artificial neural networks in predicting the stock price. Financial Engineering and Securities management(Portfolio Management), 10(39), 278-294. (in Persian)
  44. Tsai, C. F., & Hsiao, Y. C. (2010). Combining multiple feature selection methods for stock prediction: Union, intersection, and multi-intersection approaches. Decision Support Systems, 50(1), 258-269.
  45. Ul Haq, A., Zeb, A., Lei, Z., & Zhang, D. (2021). Forecasting daily stock trend using multi-filter feature selection and deep learning. Expert Systems With Applications, 168(3),114444.
  46. Van Der Maaten, L., Postma, E., & Van den Herik, J. (2009). Dimensionality reduction: A comparative. Journal of Machine Learning Research, 10(1-41), 66-71.

47.Wanga, D., & Zhao,Y.(2020). Using News to Predicton Investor Sentiment:Based on SVM Model, Procedia Computer Science, 174,191-199.

  1. Yoo, P., Kim, M., & Jan, T. (2005). Machine learning techniques and use of event information for stock market prediction: A survey and evaluation. In Proceedings - International Conference on Computational Intelligence for Modelling, Control and Automation, (CIMCA 2005),(pp. 835-841). Piscataway, NJ: IEEE.