FORECASTING THE BIST 30 INDEX SHARES USING ARTIFICIAL INTELLIGENCE TECHNIQUES

Yıl 2023, , 270 - 286, 31.03.2023

Mehmet Harun Songün , Murat Akbalık

https://doi.org/10.29106/fesa.1230607

Öz

In this article, supervised Machine Learning model was used to statistically calculate the future value of the shares in the BIST 30 index as of January 2022. The future price forecasting is calculated by using the daily opening, closing, low price, high price and volume data of the stocks in the BIST 30 index. December 2003-January 2022 was used as the daily return range. The data set covers the pandemic period that started in 2020 and the post-economic crisis period in Turkey in 2001. Different from the studies in the literature in terms of scope, two different time frame price predictions were applied to each of 30 different data sets. Data sets covers minumum 2915 and maximum 4707 trading days. Decision Treet algorithm, one of the Artificial Intelligence and Machine Learning algorithms, is used to predict the future price.

Anahtar Kelimeler

Artificial Intelligence, Decision Tree, Price Forecast

BIST 30 ENDEKSİ PAYLARININ YAPAY ZEKA YÖNTEMİYLE TAHMİNİ ÜZERİNE BİR ARAŞTIRMA

Öz

Bu makalede, 7 Ocak 2022 cuma günü itibarıyla Borsa İstanbul 30 endeksinde işlem gören payların gelecekteki değerini matematiksel model ve algoritmalarla hesaplamak için denetimli Makine Öğrenimi modeli kullanılmıştır. Gelecekteki fiyat öngörüsünü BIST 30 endeksinde yer alan payların Aralık 2003-Ocak 2022 tarihleri arasındaki günlük açılış, kapanış, düşük fiyat, yüksek fiyat ve hacim verileri kullanılarak hesaplanmıştır. Veri seti 2020 yılında başlayan pandemi dönemini ve 2001 yılında Türkiye’de yaşanan ekonomik kriz sonrası dönemi kapsamaktadır.Literatürde yer alan çalışmalardan kapsam bakımından farklı olarak her biri en az 2915 en çok 4707 işlem gününü kapsayan 30 farklı veri setine iki farklı zaman dilimi fiyat öngörüsü uygulanmıştır. Gelecek dönem fiyat öngörüsünde bulunabilmek için Yapay Zeka,Makine Öğrenimi algoritmalarından olan Karar Ağacı algoritması kullanılmıştır.

Anahtar Kelimeler

Yapay Zeka, Karar Ağac ı, Fiyat Öngörüsü, Bist30, Artificial Intelligence, Decision Tree, Price Forecast

Kaynakça

• Abe, M., Nakayama, H. (2018). Deep Learning forForecastingStockReturns in the Cross-Section. In: Phung, D., Tseng, V., Webb, G., Ho, B., Ganji, M., Rashidi, L. (eds) Advances in Knowledge Discoveryand Data Mining. PAKDD 2018. LectureNotes in ComputerScience(), vol 10937
• Ahmed, Nesreen K. , Atiya, Amir F. , Gayar, Neamat El and El-Shishiny, Hisham(2010) 'An EmpiricalComparison of Machine Learning Modelsfor Time Series Forecasting', EconometricReviews, 29: 5, 594 — 621
• Altay, E., & Satman, M. H. (2005). Stock market forecasting: artificialneural network andlinearregressioncomparison in an emerging market. Journal of Financial Management & Analysis, 18(2), 18.
• Atsalakis, G.S., &Valavanis, K.P. (2009). Surveyingstock market forecastingtechniques - Part II: Softcomputingmethods. ExpertSyst. Appl., 36, 5932-5941.
• Aziz, S., &Dowling, M. (2019). Machine learningand AI for risk management. In Disruptingfinance (pp. 33-50). Palgrave Pivot, Cham.
• Beruticha, J.M., López, F., Luna, F., Quintana, D. (2016).Robusttechnicaltradingstrategiesusing GP foralgorithmicportfolioselection. ExpertSystemswith Applications, 46, pp. 307-315.
• Bianchi, D., Büchner, M., &Tamoni, A. (2021). Bond risk premiumswithmachinelearning. TheReview of Financial Studies, 34(2), 1046-1089.
• Black, F., &Litterman, R. (1992). Global portfoliooptimization. Financial analystsjournal, 48(5), 28-43.
• Booth, A., Gerding, E., &McGroarty, F. (2015). Performance-weightedensembles of randomforestsforpredictingpriceimpact. Quantitativefinance, 15(11), 1823-1835.
• Branke, J., Scheckenbach, B., Stein, M., Deb, K., &Schmeck, H. (2009). Portfolio optimizationwith an envelope-basedmulti-objectiveevolutionaryalgorithm. EuropeanJournal of OperationalResearch, 199(3), 684-693.
• Chandar, S. K. (2022). Convolutional neural network for stock trading using technical indicators. Automated Software Engineering, 29, 1-14.
• Chavan, P. S., & Patil, S. T. (2013). Parametersforstock market prediction. International Journal of ComputerTechnologyand Applications, 4(2), 337
• Chen, A. S., &Leung, M. T. (2004). Regressionneural network forerrorcorrection in foreignexchangeforecastingandtrading. Computers& Operations Research, 31(7), 1049-1068.
• Chen, W. H., Shih, J. Y., &Wu, S. (2006). Comparison of support-vectormachinesandbackpropagationneuralnetworks in forecastingthesixmajorAsianstockmarkets. International Journal of Electronic Finance, 1(1), 49-67.
• Chong, E., Han, C., & Park, F. C. (2017). Deeplearningnetworksforstock market analysisandprediction: Methodology, datarepresentations, andcasestudies. ExpertSystemswith Applications, 83, 187-205.
• Das, S. P., &Padhy, S. (2012). Supportvectormachinesforprediction of futuresprices in Indianstock market. International Journal of Computer Applications, 41(3).
• Deniz, Ö. (2005). Poisson regresyon analizi. İstanbul Ticaret Üniversitesi Fen Bilimleri Dergisi, 4(7), 59-72.
• Donaldson, R. G., &Kamstra, M. (1997). An artificialneural network-GARCH model forinternationalstockreturnvolatility. Journal of Empirical Finance, 4(1), 17-46.
• Dondurmacı, G. A., & Çınar, A. (2014). Finans sektöründe veri madenciliği uygulaması. Akademik Sosyal Araştırmalar Dergisi, 2(1), 258-271.
• Enke, D., &Thawornwong, S. (2005). Theuse of data miningandneuralnetworksforforecastingstock market returns. ExpertSystemswithapplications, 29(4), 927- 940.
• Fan, A., &Palaniswami, M. (2001, July). Stockselectionusingsupportvectormachines. In IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No. 01CH37222) (Vol. 3, pp. 1793-1798). IEEE.
• Feng, G., Giglio, S., &Xiu, D. (2020). Tamingthefactorzoo: A test of newfactors. TheJournal of Finance, 75(3), 1327-1370.
• Fernandes, M., Medeiros, M. C., &Scharth, M. (2014). Modelingandpredictingthe CBOE market volatilityindex. Journal of Banking& Finance, 40, 1-10.
• Güdelek, M. U. (2019). Zaman serisi analiz ve tahmini: Derin öğrenme yaklaşımı (Master'sthesis, TOBB ETÜ Fen Bilimleri Enstitüsü).
• Giamouridis, D. (2017). Systematicinvestmentstrategies. Financial AnalystsJournal, 73(4), 10-14.
• Gu, S., Kelly, B., &Xiu, D. (2020). Empiricalassetpricingviamachinelearning. TheReview of Financial Studies, 33(5), 2223-2273.
• Hadavandi, E., Shavandi, H., &Ghanbari, A. (2010). Integration of geneticfuzzysystemsandartificialneuralnetworksforstockpriceforecasting. Knowledge-BasedSystems, 23(8), 800-808.
• Hu, H., Tang, L., Zhang, S., & Wang, H. (2018). Predicting the direction of stock markets using optimized neural networks with Google Trends. Neurocomputing, 285, 188-195.
• Iqbal, Z., Ilyas, R., Shahzad, W., Mahmood, Z., &Anjum, J. (2013). Efficientmachinelearningtechniquesforstock market prediction. International Journal of EngineeringResearchand Applications, 3(6), 855-867.
• Jasic, T., &Wood, D. (2004). Theprofitability of dailystock market indicestradesbased on neural network predictions: Case studyforthe S&P 500, the DAX, the TOPIX andthe FTSE in theperiod 1965–1999. Applied Financial Economics, 14(4), 285-297.
• Kim, K. J., & Han, I. (2000). Geneticalgorithmsapproachtofeaturediscretization in artificialneuralnetworksfortheprediction of stockpriceindex. Expertsystemswith Applications, 19(2), 125-132.
• Kim, K. J., & Lee, W. B. (2004). Stock market predictionusingartificialneuralnetworkswith optimal featuretransformation. Neuralcomputing&applications, 13(3), 255-260.
• Kim, M. J., Min, S. H., & Han, I. (2006). An evolutionaryapproachtothecombination of multiple classifierstopredict a stockpriceindex. ExpertSystemswith Applications, 31(2), 241-247.
• Kim, H. J., &Shin, K. S. (2007). A hybridapproachbased on neuralnetworksandgeneticalgorithmsfordetecting temporal patterns in stockmarkets. AppliedSoft Computing, 7(2), 569-576.
• Koç Ustalı, N., Tosun, N., Tosun, Ö. (2021). “Makine Öğrenmesi Teknikleri ile Hisse Senedi Fiyat Tahmini”, Eskişehir Osmangazi Üniversitesi İİBF Dergisi, 16(1), 1 – 16.
• Lee, M. C. (2009). Using supportvectormachinewith a hybridfeatureselectionmethodtothestock trend prediction. ExpertSystemswith Applications, 36(8), 10896-10904.
• Liao, Z., & Wang, J. (2010). Forecasting model of global stockindexbystochastic time effectiveneural network. ExpertSystemswith Applications, 37(1), 834-841.
• Lopez de Prado, M. (2016). BuildingDiversifiedPortfoliosThatOutperformOut-of-Sample (Presentation Slides). Available at SSRN 2713516.
• Ma, Y., Han, R., & Wang, W. (2021). Portfolio optimization with return prediction using deep learning and machine learning. Expert Systems with Applications, 165, 113973.
• Murphy, K. P. (2012). Machine learning: a probabilisticperspective. MIT press.
• Müller, K. R., Smola, A. J., Rätsch, G., Schölkopf, B., Kohlmorgen, J., &Vapnik, V. (1997, October). Predicting time serieswithsupportvectormachines. In International conference on artificialneuralnetworks (pp. 999-1004). Springer, Berlin, Heidelberg.
• Nabipour, M., Nayyeri, P., Jabani, H., Shahab, S., & Mosavi, A. (2020). Predicting stock market trends using machine learning and deep learning algorithms via continuous and binary data; a comparative analysis. IEEE Access, 8, 150199-150212.
• Oh, K. J., & Han, I. (2000). Using change-pointdetectiontosupportartificialneuralnetworksforinterestratesforecasting. Expertsystemswithapplications, 19(2), 105-115.
• Ou, P., & Wang, H. (2009). Prediction of stock market indexmovementby ten data miningtechniques. Modern AppliedScience, 3(12), 28-42.
• Pierdzioch, C., &Risse, M. (2018). A machine‐learninganalysis of therationality of aggregatestock market forecasts. International Journal of Finance &Economics, 23(4), 642-654.
• Rasekhschaffe, K. C., & Jones, R. C. (2019). Machine learningforstockselection. Financial AnalystsJournal, 75(3), 70-88.
• Rapach, D. E., Strauss, J. K., Tu, J., & Zhou, G. (2019). Industryreturnpredictability: A machinelearningapproach. TheJournal of Financial Data Science, 1(3), 9-28.
• Sabharwal, C. L. (2018). Therise of machinelearningandrobo-advisors in banking. IDRBT Journal of BankingTechnology, 28.
• Skolpadungket, P., Dahal, K., &Harnpornchai, N. (2016). Handling Model Risk in Portfolio Selection Using Multi-ObjectiveGeneticAlgorithm. In ArtificialIntelligence in Financial Markets (pp. 285-310). Palgrave Macmillan, London.
• Schumaker, R. P., & Chen, H. (2009). Textualanalysis of stock market predictionusingbreakingfinancial news: TheAZFintextsystem. ACM Transactions on Information Systems (TOIS), 27(2), 12.
• Strader, Troy J.; Rozycki, John J.; ROOT, THOMAS H.; and Huang, Yu-Hsiang (John) (2020) "Machine Learning Stock Market PredictionStudies: ReviewandResearchDirections," Journal of International Technologyand Information Management: Vol. 28 : Iss. 4 , Article 3.
• ŞIKLAR, E. (1999) REGRESYON ANALİZİNDE BAYESCİ YAKLAŞIM. Anadolu Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 15(1), 113-122.
• Tan, P. N., Steinbach, M., & Kumar, V. (2013). Data miningclusteranalysis: basicconceptsandalgorithms. Introductionto data mining, 487, 533.
• Tunçel, A. K. (2007). Rassal yürüyüş (randomwalk) hipotezi’nin İMKB’de test edilmesi: koşu testi uygulamasi. Gazi Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 9(2), 1-18.
• Urquhart, A., Gebka, B., & Hudson, R. (2015). How exactly do marketsadapt? Evidencefromthemovingaveragerule in threedevelopedmarkets. Journal of International Financial Markets, Institutionsand Money, 38, 127-147.
• Viswanathan, K., Choudur, L., Talwar, V., Wang, C., Macdonald, G., &Satterfield, W. (2012, April). Rankinganomalies in data centers. In 2012 IEEE Network Operations and Management Symposium (pp. 79-87). IEEE.
• Vochozka, M., &Sheng, P. (2016). Theapplication of artificialneuralnetworks on theprediction of thefuturefinancialdevelopment of transport companies. Communications-Scientificletters of theUniversity of Zilina, 18(2), 62-67.
• Wang, Y., Wong, J., &Miner, A. (2004, June). Anomalyintrusiondetectionusingoneclass SVM. In ProceedingsfromtheFifthAnnual IEEE SMC Information Assurance Workshop, 2004. (pp. 358-364). IEEE.
• Xiao-si, X., Ying, C., &Ruo-en, R. (2006, October). Studying on forecastingtheenterprisebankruptcybased on SVM. In 2006 International Conference on Management ScienceandEngineering (pp. 1041-1045). IEEE.
• Yeh, C. Y., Huang, C. W., & Lee, S. J. (2011). A multiple-kernelsupportvectorregressionapproachforstock market priceforecasting. ExpertSystemswith Applications, 38(3), 2177-2186.
• Yu, L., Chen, H., Wang, S., &Lai, K. K. (2008). Evolvingleastsquaressupportvectormachinesforstock market trend mining. IEEE Transactions on evolutionarycomputation, 13(1), 87-102.
• Zhong, X., & Enke, D. (2019). Predicting the daily return direction of the stock market using hybrid machine learning algorithms. Financial Innovation, 5(1), 1-20.
• Zimmermann, H. G., Neuneier, R., &Grothmann, R. (2001). Active portfolio-managementbased on errorcorrectionneuralnetworks. Advances in Neural Information ProcessingSystems, 14.