Journals / Ankara Hacı Bayram Veli Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi / 2022 / Cilt: 24 - Sayı: 3

Financial Time Series Prediction Using Machine Learning

Makine Öğrenmesi ile Finansal Zaman Serisi Tahminleme

Pages
1205–1230
Publication Date
2022-12-28
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

As a result of the study, when MAE, MAPE and RMSE criterions are taken into consideration, SVR was confirmed to be the best forecasting method.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Making up an important working area of finance applications, financial time series forecasting, with the advancements in Machine Learning (ML) methods in recent years, has become a topic that finance and academic circles attach more importance to.
📊
Key Findings & Data:In the study, the last 5-year closing data of two developed and emerging stock market indices and two high-volume stocks of the Istanbul stock market were used.
💡
Conclusion & Impact:As a result of the study, when MAE, MAPE and RMSE criterions are taken into consideration, SVR was confirmed to be the best forecasting method.
🔬Design:Review & Systematic Survey
📍Scope:Comprehensive Sample
⚙️Variables:finansal zaman serisi · makine öğrenmesi · destek vektör makinesi

Abstract

Making up an important working area of finance applications, financial time series forecasting, with the advancements in Machine Learning (ML) methods in recent years, has become a topic that finance and academic circles attach more importance to. The aim of this study is to present a comparative review of ML methods in financial time series future value. In the study, the last 5-year closing data of two developed and emerging stock market indices and two high-volume stocks of the Istanbul stock market were used. Support Vector Regression (SVR), which is often used in index forecasting and found successful, and Random Forest (RF) and eXtreme Gradient Boosting (XGB) methods which are rarely used ensemble machine learning methods in time series forecasting in literature, are preferred. As a result of the study, when MAE, MAPE and RMSE criterions are taken into consideration, SVR was confirmed to be the best forecasting method.

Özet

Finans uygulamalarının önemli bir çalışma alanını oluşturan finansal zaman serisi tahminlemesi son yıllarda makine öğrenmesi (Machine Learning, ML) yöntemlerinin gelişimi ile finans ve akademi çevrelerinin daha fazla önem atfettiği bir konu olmuştur. Bu çalışmanın amacı, finansal zaman serisi gelecek değerinin tahmininde ML yöntemlerinin karşılaştırmalı olarak bir incelemesini sunmaktır. Çalışmada gelişmiş ve gelişmekte olan iki borsa endeksi ve İstanbul borsasının yüksek hacimli iki hisse senedinin son 5 yıllık kapanış verileri kullanılmıştır. Endeks tahmininde sıklıkla kullanılmış ve başarılı bulunan Destek Vektör Regresyonu (Suport Vector Regression, SVR) ve literatürde zaman serisi tahmininde izine az rastladığımız topluluk (ensemble) makine öğrenmesi yöntemleri olan Rassal Orman (Random Forest, RF) ve Extrem Gradyan Arttırma (eXtreme Gradient Boosting, XGB) yöntemleri tercih edilmiştir. Çalışmanın bulgularına göre, MAE, MAPE ve RMSE kriterleri göz önünde bulundurulduğunda en iyi tahmin yöntemi SVR olarak tespit edilmiştir.

🔖

Cite This Article

DOĞAN, S., BÜYÜKKÖR, Y. (2022). Financial Time Series Prediction Using Machine Learning. Ankara Hacı Bayram Veli Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, Cilt: 24 - Sayı: 3, 1205–1230. https://ajindex.com/articles/877268

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (11)

  1. Abraham, A., Nath, B., & Mahanti, P. K. (2001, May). Hybrid İntelligent Systems for Stock Market Analysis. In International Conference on Computational Science (pp. 337-345),
  2. Springer, Berlin, Heidelberg. Ahmad, M.W., Reynolds, J., Rezgui, Y. (2018). Predicti& Modelling for Solar Thermal Energy Systems: A Comparison of Support Vector Regression, Random Forest, Extra Trees And Regression Trees, Journal of Cleaner Production, 203, 810-821.
  3. Akita, R., Yoshihara, A., Matsubara, T., & Uehara, K. (2016, June). Deep Learning for Stock Prediction Using Numerical and Textual İnformation. In 2016 IEEE/ACIS 15th International Conference on Computer and Information Science (ICIS) (pp. 1-6).
  4. IEEE. Arslankaya, S., & Toprak, Ş. (2021). Makine Öğrenmesi ve Derin Öğrenme Algoritmalarını Kullanarak Hisse Senedi Fiyat Tahmini. International Journal of Engineering Research and Development, 13(1), 178-192.
  5. Ashfaq, N., Nawaz, Z., & Ilyas, M. (2021). A Comparative Study of Different Machine Learning Regressors for Stock Market Prediction. Arxiv Preprint Arxiv:2104.07469. Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5-32.
  6. Cao, J., & Wang, J. (2019). Stock Price Forecasting Model Based on Modified Convolution Neural Network and Financial Time Series Analysis. International Journal of Communication Systems, 32(12), e3987. Cao, J., Li, Z., & Li, J. (2019). Financial Time Series Forecasting Model Based On CEEMDAN And LSTM. Physica A: Statistical Mechanics and Its Applications, 519, 127-139.
  7. Cao, L. J., & Tay, F. E. H. (2003). Support Vector Machine with Adaptive Parameters in Financial Time Series Forecasting. IEEE Transactions on Neural Networks, 14, 1506– 1518. Doi:10.1109/TNN.2003.820556. Cao, L., & Tay, F. E. (2001). Financial Forecasting Using Support Vector Machines. Neural Computing and Applications, 10(2), 184-192.
  8. Chen SM (1996) Forecasting Enrollments Based On Fuzzy Time-Series. Fuzzy Sets Syst 81:311–319.
  9. Chen, H., Xiao, K., Sun, J., & Wu, S. (2017). A Double-Layer Neural Network Framework for High-Frequency Forecasting. ACM Transactions on Management Information Systems (TMIS), 7(4), 1-17.
  10. Chen, T., & Guestrin, C. (2016, August). Xgboost: A Scalable Tree Boosting System. In Proceedings of The 22nd Acm Sigkdd İnternational Conference on Knowledge Discovery and Data Mining (pp. 785-794).
  11. Chen, Y. S., Cheng, C. H., & Tsai, W. L. (2014). Modeling Fitting-Function-Based Fuzzy Time Series Patterns for Evolving Stock İndex Forecasting. Applied İntelligence, 41(2)