COMPARISON OF CLASSIFICATION PERFORMANCE OF MACHINE LEARNING METHODS IN PREDICTION FINANCIAL FAILURE: EVIDENCE FROM BORSA İSTANBUL

Abstract

This study aimed to predict the 1 to 2 year future time of the financial failure of 86 manufacturing companies that are operating in Borsa İstanbul. The data comprised of 2010-2012 period, and it depends on 8 quantitative financial variables. Beside 6 variables come from non financial statements. In the study, Artificial Neural Network (NN), Classification and Regression Trees (CART), Support Vector Machine (SVM) and k-Nearest Neighbors (KNN) were used to compare classification performances of related methods. ROC Curve was used to compare the classification performance of the methods. As a result of the analyseis, the overall classification accuracy from the highest to the lowest was SVM (92,31%), CART (88,46%), ANN (84,62%) and KNN (80,77%) 2 years before the financial failure. The overall classification accuracy from the highest to the lowest was CART (96,15%), ANN (92,31%), SVM (80,77%) and KNN (84,62%) 1 year before the financial failure. Return on Equity (ROE) and Return on Assets Ratio (ROA) were found as important variables in the creation of the CART decision tree. The fact that the four models obtained in thise study predicted financial success/failure at a higher rate, and it shows that the models obtained in this study can be included in the models used by relevant people.

Keywords

• Financial Failure Prediction
• Borsa İstanbul
• Artificial Neural Networks
• Classification and Regression Trees
• Support Vector Machine
• KNN Algorithm

FİNANSAL BAŞARISIZLIK TAHMİNİNDE MAKİNE ÖĞRENMESİ YÖNTEMLERİNİN SINIFLANDIRMA PERFORMANSININ KARŞILAŞTIRILMASI: BORSA İSTANBUL ÖRNEĞİ

Abstract

Bu çalışmada Borsa İstanbul İmalat Sanayi Sektörüne kayıtlı 86 firmanın, 2010-2012 dönemine ait verileri kullanılarak 1 ve 2 yıl öncesinden finansal başarısızlık tahmini yapılmıştır. Araştırmada 8 mali tablolara dayalı nicel ve 6 mali tablolara dayalı olmayan değişken kullanılmıştır. Çalışma amacına yönelik analizlerde Yapay Sinir Ağları (ANN), Sınıflandırma ve Regresyon Ağaçları (CART), Destek Vektör Makineleri (SVM) ve K-En Yakın Komşular Algoritması (KNN) yöntemlerinin tahmin performansları yöntemlerin ayırt edici özellikleri altında karşılaştırılmıştır. ROC Eğrisi yöntemlerin sınıflandırma performanslarını karşılaştırmak için kullanılmıştır. Analiz sonucunda, finansal başarısızlıktan iki yıl önce en yüksekten düşüğe genel sınıflandırma doğruluğu SVM (% 92,31), CART (%88,46), ANN (% 84,62), KNN (%80,77) olarak bulunmuştur. Finansal başarısızlıktan bir yıl önce en yüksekten en düşüğe genel sınıflandırma doğruluğu CART (% 96,15), ANN (%92,31), SVM (% 80,77) ve KNN (%84,62) olarak elde edilmiştir. CART karar ağacının oluşturulmasında önemli değişkenler olarak Özsermaye kârlılığı (ROE) ve Aktif Kârlılık Oranı (ROA) bulunmuştur. Çalışmada elde edilen dört modelin finansal başarı/başarısızlığı bir ve iki yıl öncesinden yüksek oranda tahmin etmesi, ilgililerin kullandıkları modeller içerisine bu çalışmada elde edilen modelleri dâhil edebileceklerini göstermektedir.

Keywords

• Finansal Başarısızlık Tahmini
• Borsa Istanbul
• Yapay Sinir Ağları
• Sınıflandırma ve Regresyon Ağaçları
• Destek Vektör Makinesi,
• KNN Algoritması

References

1. Akay, E. Ç. (2018), A New horizon in econometrics: big data and machine learning. Social Sciences Research Journal, 7(2), 41-53. Retrieved from: https://dergipark.org.tr/tr/pub/ssrj/issue/37241/423147
2. Akkaya, G. C., Demireli, E. and Yakut, Ü. H. (2009). Using artificial neural networks in financial failure prediction: an application in Borsa Istanbul. Journal of Social Sciences, Eskişehir Osmangazi University, 10(2), 187-216. Retrieved from: https://dergipark.org.tr/tr/pub/ogusbd/issue/10996/131592
3. Akpınar, H. (2014). Data mining data analysis. İstanbul: Papatya Publications.
4. Alfaro, E., Garcia, M. G. N. & Elizondo, D. (2008). Bankruptcy forecasting: an empirical comparison of adaboost and neural networks. Decision Support Systems, 45, 110-122. doi: 10.1016/j.dss.2007.12.002
5. Bilir, H. (2015). Definition and Market oriented solution of financial distress: debt structuring, asset sales and new capital injection. Sosyoekonomi,1, 9-24. Retrieved from: https://dergipark.org.tr/tr/download/article-file/197806
6. Chandra, D. K., Ravi, V. & Bose, I. (2009). Failure prediction of DOTCOM companies using hybrid intelligent techniques. Expert Systems with Applications, 36, 4830-4837. doi: 10.1016/j.eswa.2008.05.047
7. Chen, Mu-Y. (2011). Predicting corporate financial distress based on integration of decision tree classification and logistic regression. Expert Systems with Applications, 38, 11261-11272. doi: 10.1016/j.eswa.2011.02.173
8. Chen, M.-Y. (2011). Bankruptcy prediction in firms with statistical and intelligent techniques and a comparison of evolutionary computation approaches. Computers and Mathematics with Applications, 62(12), 4514-4524. doi: 10.1016/j.camwa.2011.10.030

9. Chuang, C.-L. (2013). Application of hybrid case-based reasoning for enhanced performance in bankruptcy prediction. Information Sciences, 236, 174-185. doi: 10.1016/j.ins.2013.02.015
10. Çelik U., Akçetin, E. & Gök, M. (2017). Data mining with Rapidminer, İstanbul: Pusula Publications.
11. Çöllü, D. A., Akgün, L. and Eyduran, E. (2020). Financial failure prediction with decision tree algorithms: textile, wearing apparel and leather sector application. International Journal of Economics and Innovation, 6(2), 225-246. doi: 10.20979/ueyd.698738
12. Dener, M., Dörterler, M. ve Orman, A. (2009). Conference: XI. Academic Informatics Conference, Open Source Data Mining Programs: Sample Application in Weka. (p. 1-11). Şanlıurfa
13. Divsalar, M., Firouzabadi, A. K., Sadeghi, M.; Behrooz, A. H. & Alavi, A. H. (2011). Towards the prediction of business failure via computational intelligence techniques. Expert Systems, 28(3), 209-226. doi: 10.1111/j.1468-0394.2011.00580.x
14. Enke, D., & Thawornwong, S. (2005). The use of data mining and neural networks for forecasting stock market returns. Expert Systems with Applications, (29), 927-940. doi:10.1016/j.eswa.2005.06.024
15. Elmas, Ç. (2018). Artificial intelligence applications. Ankara: Seçkin Publishing.
16. Etemadi, H., Rostamy, A. A. A. & Dehkordi, H. F. (2009). A Genetic programming model for bankruptcy prediction: Empirical evidence from Iran. Expert Systems with Applications, 36, 3199–3207. doi: 10.1016/j.eswa.2008.01.012
17. Gaganis, C. (2009). Classification techniques for the identification of falsified financial statements: A comparative analysis. Intelligent Systems in Accounting, Finance and Management, (16), 207-229. doi: 10.1002/isaf.303
18. 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, 236-247. doi: 10.1016/j.ejor.2014.08.016
19. Gepp, A., Kumar, K. & Bhattacharya, S. (2010). Business failure prediction using decision trees. Journal of Forecasting, 29, 536-555. doi: 10.1002/for.1153
20. Gepp, A. & Kumar, K. (2015). Predicting financial distress: A comparison of survival analysis and decision tree techniques. Procedia Computer Science, 54, 396-404. doi: 10.1016/j.procs.2015.06.046
21. Huang, Z., Chen, H., Hsu, C-J., Chen, W.-H. & Wu, S. (2004). Credit rating analysis with support vector machines and neural networks: A market comparative study. Decision Support Systems, 37, 543-558. doi: 10.1016/S0167-9236(03)00086-1
22. Jardin, P. D. (2010). Predicting bankruptcy using neural networks and other classification methods: the influence of variable selection techniques on model accuracy. Neurocomputing, 73, 2047-2060. doi: 10.1016/j.neucom.2009.11.034
23. Jardin, P. D. (2012). The inflcuence of variable selection methods on the accuracy of bankruptcy prediction models. Bankers, Markets & Investors,116, 20-39. Retrieved from:https:www.researchgate.net/publication/235643793
24. Jardin, P. D. & Séverin, E. (2011). Predicting corporate bankruptcy using a self-organizing map: An empirical study to ımprove the forecasting horizon of a financial failure model. Decision Support Systems, 51, 701-711. doi: 10.1016/j.dss.2011.04.001
25. Jardin, P. D. (2016). A Two-Stage classification technique for bankruptcy prediction. European Journal of Operational Research, 254, 236-252. doi: 10.1016/j.ejor.2016.03.008
26. Kara, Y., Boyacıoğlu, M. A., & Baykan, Ö. K. (2011). Predicting direction of stock price index movement using artificial neural networks and support vector machines: the sample of the İstanbul Stock Exchange. Expert Systems with Applications, (38), 5311-5319. doi: 10.1016/j.eswa.2010.10.027
27. Kaygın, C. Y., Tazegül, A. and Yazarkan, H. (2016). Estimation capability of financial failures and successes of enterprises using data mining and logistic regression analysis. Ege Academic Review, 16(1), 147-159. doi: 10.21121/eab.2016116590
28. Keasey, K. & Watson, R. (1987). Non-financial symptoms and the prediction of small company failure: a test of arcenti’s hypotheses. Journal of Business Finance and Accounting, 14(3), 335-354. doi: 10.1111/J.1468-5957.1987.TB00099.X
29. Kılıç, S., (2015). Kappa Test. Journal of Mood Disorders, 5(3), 142-144. doi: 10.5455/jmood.20150920115439
30. Koç, S. and Ulucan S. (2016). Testing of Altman Z methods which is used for detecting of financial failures with fuzzy logic (Anfis) technique: a case study on technology and textile sector. Journal of Finance Letters, (106), p. 147-167. doi: 10.33203/mfy.341768
31. Le, H. H. and Viviani, J.-L. (2018). Predicting bank failure: An improvement by implementing a machine-learning approach to classical financial ratios. Research in International Business and Finance, (44), 16-25. doi: 10.1016/j.ribaf.2017.07.104
32. Li, H. & Sun, J. (2009). Predicting business failure using multiple case-based reasoning combined with support vector machine. Expert Systems with Applications, 36, 10085–10096. doi: 10.1016/j.eswa.2009.01.013
33. Li, H., Sun, J. & Wu, J. (2010). Predicting business failure using classification and regression tree: an empirical comparison with popular classical statistical methods and top classification mining methods. Expert Systems with Applications, 37, 5895-5904. doi: 10.1016/j.eswa.2010.02.016
34. Li, H. and Sun, J. (2013). Predicting business failure using an rsf-based case-based reasoning ensemble forecasting method. Journal of Forecasting, (32), 180-192. doi: 10.1002/for.1265
35. Li, A. & Wu J. and Zhidong L. (2017). Market manipulation detection based on classification methods. Procedia Computer Science, 122; 788-795. doi: 10.1016/j.procs.2017.11.438
36. Liang, D., Tsaı, C.-F. & Wu, H.-T. (2015). The effect of feature selection on financial distress prediction. Knowledge-Based Systems, 73, 289-297. doi: 10.1016/j.knosys.2014.10.010
37. Lin, F., Liang, D., Yeh, C.-C. & Huang, J.-C. (2014). Novel feature selection methods to financial distress prediction. Expert Systems with Applications, 41, 2472–2483. doi: 10.1016/j.eswa.2013.09.047
38. Min, J. H. & Lee, Y.-C. (2005). Bankruptcy prediction using support vector machine with optimal choice of kernel function parameters. Expert Systems with Applications, 28, 603–614. doi: 10.1016/j.eswa.2004.12.008
39. Muller, G.H, Steyn-Bruwer, B. W & Hamman, W.D. (2009). Predicting financial distress of companies listed on the JSE- A comparison of techniques. S. Afr.Bus. Manage, 40(1), 21-32. doi: 10.4102/sajbm.v40i1.532
40. Özçalıcı, M. (2017). Stock price prediction with extreme learning machines. Hacettepe University Journal of Economics and Administrative Sciences, 35(1), 67-88. doi: 10.17065/huniibf.303305
41. Özdağoğlu, G., Özdağoğlu, A., Gümüş, Y. & Gümüş, G. K. (2017). The application of data mining techniques in manipulated financial statement classification: The case of Turkey. Journal of AI and Data Mining, 5(1), 67-77. doi: 10.22044/jadm.2016.664
42. Özkan, Y. (2016). Data mining methods. İstanbul: Papatya Publications.
43. Öztemel, E. (2012). Artificial neural networks, İstanbul: Papatya Publications.
44. Sayılgan, G. and Ece, A. (2016). Bankruptcy postponement and panoramic analysis on bankruptcy postponement litigations in Turkey between 2009-2013. Revenue and Finance Articles, (105), 47-74. doi: 10.33203/mfy.312132
45. Selimoglu, S. & Orhan, A. (2015). Measuring business failure by using ratio analysis and discriminant analysis: a research on textile, clothes and leather firms listed in the İstanbul Stock Exchange. Journal of Accounting and Finance, April, 21-40. doi: 10.25095/mufad.396529
46. Shin, K.-S., Lee, T. S & Kim, H.-j. (2005). An application of support vector machines in bankruptcy prediction model. Expert Systems with Applications, 28, 127-135. doi: 10.1016/j.eswa.2004.08.009
47. Silahtaroğlu, G. (2016). Data mining concepts and algorithms. İstanbul: Papatya Publications.
48. Sun, J., Li, H., Huang, Q.-H. & He, K.-Y. (2014). Predicting financial distress and corporate failure: a review from the state-of-the-art definitions, modeling, sampling, and featuring approaches. Knowledge-Based Systems, 57, 41-56. doi: 10.1016/j.knosys.2013.12.006
49. Tsai, C. F., Hsu, Y. F. and Yen D. C. (2014). A comparative study of classifier ensembles for bankruptcy prediction. Applied Soft Computing, (24), 977-984. doi: 10.1016/j.asoc.2014.08.047
50. Wu, W., Lee, V. C. S. & Tan, T. Y. (2006). Data preprocessing and data parsimony in corporate failure forecast models: Evidence from Australian materials industry. Accounting and Finance, 46, 327-345. doi: 10.1111/j.1467-629X.2006.00170.x
51. Yakut, E. (2012). The Comparison of the classification successes of the artifical neural networks through data mining techniques of c5.0 algorithm and supporting vector machines: an application in manufacturing sector, Atatürk University Social Sciences Institute, Erzurum (Unpublished PhD Thesis). Access address: https://tezarsivi.com/veri-madenciligi-tekniklerinden-c50-algoritmasi-ve-destek-vektor-makineleri-ile-yapay-sinir-aglarinin-siniflandirma-basarilarinin-karsilastirilmasi-imalat-sektorunde-bir-uygulama
52. Yürük, M. F. and Ekşi, İ. H. (2019). Financial failure prediction of companies using artificial i̇ntelligence methods: An application in Bist manufacturing sector. Mukaddime, 10(1), 393-422. doi: 10.19059/mukaddime.533151