İŞLETMELERDE FİNANSAL BAŞARISIZLIK ÖNGÖRÜSÜNDE MAKİNE ÖĞRENMESİNİN KULLANIMI: BİST UYGULAMASI

Öz

Çalışmanın amacı işletmelerin finansal başarısızlık riski ile ilgili tahmin yapay zekâ tekniklerinden makine öğrenmesi kullanılarak yapılmasıdır. Bu kapsamda, Borsa İstanbul Ulusal Pazar’da yer alan 14 firma ile Borsa İstanbul Yakın İzleme Pazarı’nda yer alan14 firmanın 2022 yılı 12 aylık gelir tabloları ve bilançolarından elde edilen 43 adet finansal oran kullanılmış makine öğrenmesi yöntemlerinden NaiveBayes, J48, RandomForest, LinearRegression, RandomTree kullanılmıştır. Şirketlerin mali tabloları kullanılarak elde edilen veriler ile, makine öğrenmesi uygulama modellerinden hangisinin daha iyi sınıflandırma doğruluğu sağladığı araştırılmıştır. Ayrıca 2022 yılında yakın izleme pazarında yer alan bir şirketin bir sonraki sene için finansal durumunun makine öğrenmesi ile öngörüsünün mümkün olup olmadığı test edilmiştir. En yüksek sınıflandırma doğruluğu oranına RandomForest algoritması ve 10 kat çapraz doğrulama tekniğinin birlikte uygulanması ile ulaşıldığı, tek yıl için yapılan öngörü modelinde ise NaiveBayes algoritması ve 10 kat çapraz doğrulama tekniğinin çok yüksek bir oranda başarı sağladığı sonuçlarına ulaşılmıştır.

Anahtar Kelimeler

• Sınıflandırma
• Finansal Sıkıntı
• Finansal Başarısızlık
• Makine Öğrenmesi
• Tahmin

THE USE OF MACHINE LEARNING IN PREDICTING FINANCIAL FAILURE IN BUSINESSES: ISTANBUL STOCK EXCHANGE APPLICATION

Öz

The aim of the study is to estimate the risk of financial failure of enterprises by using machine learning, one of the artificial intelligence techniques. In this context, machine learning methods NaiveBayes, J48, RandomForest, LinearRegression, RandomTree were used with 43 financial ratios obtained from the financial statements of 14 companies in Borsa Istanbul Main Market and 14 companies in Borsa Istanbul Watchlist Market for the year 2022. With the data obtained using the financial statements of the companies, it is investigated which of the machine learning application models provides better classification accuracy. In addition, it was tested whether it is possible to predict the financial situation of a company in the close monitoring market in 2022 for the following year with machine learning. It was concluded that the highest classification accuracy rate was achieved by applying the RandomForest algorithm and 10-fold cross-validation technique together, while the NaiveBayes algorithm and 10-fold cross-validation technique achieved a very high rate of success in the prediction model for a single year.

Anahtar Kelimeler

• Classification
• Financial Distress
• Financial Failure
• Machine Learning
• Prediction

Kaynakça

1. AIX, (2023, 9, 2). https://aix.web.tr/en/yapay-zeka-ve-veri-madenciligi/ adresinden alındı
2. ALAKA, H.A., OYEDELE, L.O., OWOLABİ, H.A., KUMAR, V., AJAYİ, S.O., AKİNADE, O.O., & BİLAL, M. (2018). Systematic review of bankruptcy prediction models: Towards a framework for tool selection. Expert Syst. Appl., 94, 164–184.
3. ALBAYRAK, A., & KOLTAN YILMAZ, Ş. (2009). Veri Madenciliği Karar Ağacı Algoritmaları ve İMKB Verileri Üzerine Bir Uygulama. Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 14(1), 31-52.
4. ALPAYDIN, E. (2013). Yapay Öğrenme: Introduction to Machine Learning. İstanbul: Boğaziçi Üniversitesi Yayınevi.
5. ALTMAN, E.I. (1968). Financial Ratios, Discriminant analysis and the prediction of corporate bankruptcy. J. Financ. 23, 589–609.
6. BALCAEN, S., & OOGHE, H. (2006). 35 years of studies on business failure: An overview of the classic statistical methodologies and their related problems. Br. Account. Rev., 38, 63–93.
7. BARBOZA, F., KİMURA, H., & ALTMAN, E. (2017). Machine learning models and bankruptcy prediction. Expert Syst. Appl., 83, 405–417.
8. BEAVER, W.H. (1966). Financial ratios as predictors of failure. J. Account. Res. 4, 71–111.

9. BELL, J. (2015). Machine Learning: Hands-On for Developers and Technical Professionals. New Jersey: John Wiley & Sons.
10. BİST Şirketleri. (2023, 7, 2). Kamu Aydınlatma Platformu: http://www.kap.gov.tr/sirketler/islem-goren-sirketler/pazarlar adresinden alındı
11. BORAN, L. (2012). Veri Madenciliğinin Türk İşletmelerin Finansal Tablolarına Uygulanması ve Uygulama Örneği. Yayınlanmamış Doktora Tezi, Marmara Üniversitesi S.B.E., İstanbul.
12. BROGAARD, J., LI, D., & XIA, Y. (2017). Stock lIquidity and default risk. J. Financ. Econ., 124, 486–502.
13. CHEN, W. S., & DU, Y. K. (2009). Using Neural Networks and Data Mining Techniques For The Financial Distress Prediction Model. Expert systems with applications, 36(2), 4075-4086.
14. CEYHAN, İ. F. (2014). Bağımsız denetim kalitesini artırıcı bir yöntem olarak veri madenciliği: Borsa İstanbul uygulaması. Yayımlanmamış Doktora Tezi, Sakarya Üniversitesi Sosyal Bilimler Enstitüsü, Sakarya.
15. ÇABUK, A., & LAZOL, İ. (2011). Mali Tablolar Analizi (11. b.), Bursa: Ekin Yayınevi.
16. DEMİRCİ, F. (2023). Finansta Yapay Zekâ ve Makine Öğrenme Üzerine Bibliyometrik Bir Araştırma. E. B. Ceyhan, & İ. F. Ceyhan içinde, Yapay Zekâ Alan Uygulamaları-1 (s. 1-25). Ankara: Nobel.
17. DEVI, S.S., & RADHIKA, Y. A (2018). A survey on machine learning and statistical techniques in bankruptcy prediction. Int. J. Mach. Learn. Comput., 8, 133–139.
18. DOMINGOS, P. & PAZZANI, M. (1997). On the optimality of the simple Bayesian classifier under zero-one loss. Machine Learning, 29, 103–130.
19. FAROOQ, U., & QAMAR, M. A. J. (2019). Predicting Multistage Financial Distress: Reflections on Sampling, Feature and Model Selection Criteria. Journal of Forecasting, 38(7), 632-648.
20. GOOD, I.J. (1951). Probability and the Weighing of Evidence, Philosophy, 26, 97, 163-164. https://doi.org/10.1017/S0031819100026863.
21. GENG, R., BOSE, I., & CHEN, X. (2015). Prediction of Financial Distress: An Empirical Study of Listed Chinese Companies Using Data Mining. European Journal of Operational Research, 241(1), 236-247.
22. GLOVER, B. (2016). The expected cost of default. J. Financ. Econ., 119, 284–299.
23. HASTIE, T., TIBSHIRANI, R., & FRIEDMAN, J. H. (2001). The elements of statistical learning, Data mining, inference, and prediction, New York: Springer Verlag.
24. HILLEGEIST, S.A., KEATING, E.K., CRAM, D.P., & LUNDSTEDT, K.G. (2004). Assessing the probability of bankruptcy. Rev. Account. Stud., 9, 5–34.
25. JESSEN, C., & LANDO, D. (2015). Robustness of distance-to-default. J. Bank. Financ., 50, 493–505.
26. İSTATİSTİK SİTESİ. (2023, 5, 9). http://www.stat.gen.tr/index.php?istek=sinif&dersid=ist01&konuid= ver01&max=1 adresinden alındı
27. KHEMAKHEM, S., & BOUJELBENE, Y. (2018), Predicting Credit Risk on The Basis of Financial and Non-Financial Variables and Data Mining, Review of Accounting and Finance, 17 (3), 316-340.
28. KOTSIANTIS, S. B. (2007). Supervised Machine Learning: A Review of Classification Techniques. Informatica, 31, 249 – 268.
29. KOU, G., PENG, Y., & GUOXUN, W. (2014). Evaluation of Clustering Algorithms for Financial Risk Analysis Using MCDM Methods. Information Sciences, 275, 1-12.
30. KULALI, İ. (2014). Muhasebe Temelli Tahmin Modelleri Işığında, Finansal Sıkıntı ve İflasın Karşılaştırılması. Sosyoekonomi Dergisi, 22, 153-170.
31. KUZEY, C., UYAR, A., & DELEN, D. (2014). The Impact of Multinationality on Firm Value: A Comparative Analysis of Machine Learning Techniques. Decision Support Systems, 59, 127–142.
32. LOJİSTİK REGRESYON, (2023, 8,10). 234 – 262. https://www.stat.cmu.edu/~cshalizi/ADAfaEPoV/ADAfaEPoV.pdf,
33. MCKEE, T. (1989). Modern Analytical Auditing: Practical Guidance for Auditors and Accountants. New York: Quorum Books.
34. MERTON, R.C. (1974). On the pricing of corporate debt: The risk structure of interest rates. J. Financ., 29, 449–470.
35. MITCHELL, T. (1997). Machine Learning, New York: McGraw Hill.
36. NEWSOM, I. (2015). Data Analysis II: Logistic Regression. http://web.pdx.edu/~newsomj/da2/ho_logistic.pdf
37. NILSSON, N.J. (1965). Learning machines. New York: McGraw-Hill.Published in: Journal of IEEE Transactions on Information Theory 12(3). doi: 10.1109/TIT.1966.1053912
38. NISBET, R., ELDER, J., & MINER, G. (2009). Handbook of Statistical Analysis and Data Mining Applications, San Diego: Academic Press.
39. OHLSON, J.A. (1980). Financial ratios and the probabilistic prediction of bankruptcy. J. Account. Res., 18, 109–131.
40. PATİL, R., & BARKADE, V. M. (2018). Class-Specific Features Using J48 Classifier for Text Classification, 2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA), Pune, India, 1-5, doi: 10.1109/ICCUBEA.2018.8697473.
41. PARA LİMANI. (2023, 3, 7). http://www.paralimani.com/gozalti-pazari-ne-demek-haberi-38186/ adresinden alındı
42. RAVI KUMAR, P., & RAVI, V. (2007). Bankruptcy prediction in banks and firms via statistical and intelligent techniques—A review. Eur. J. Oper. Res., 180, 1–28.
43. SAMUEL, A.L. (1959). Some studies in machine learning using the game of checkers. IBM J. Res. Dev., 3, 210–229.
44. SALEHI, M., MOUSAVI SHIRI, M., & BOLANDRAFTAR PASIKHANI, M. (2016). Predicting Corporate Financial Distress Using Data Mining Techniques: An Application in Tehran Stock Exchange, International Journal of Law and Management, 58 (2) 216-230.
45. SHANG, H., LU, D., & ZHOU, Q. (2021). Early Warning of Enterprise Finance Risk of Big Data Mining in Internet of Things Based on Fuzzy Association Rules. Neural Comput & Applic, 33, 3901–3909.
46. TAIWO, O. A. (2010). Types of Machine Learning Algorithms, New Advances in Machine Learning, Yagang
47. Zhang (Ed.), University of Portsmouth United Kingdom. 3 – 31. VAPNIK, V. N. (1995). The Nature of Statistical Learning Theory. (2nd ed.). New York: Springer Verlag. 1–20.