SavedAdd your journal

Journals / Anadolu Üniversitesi Sosyal Bilimler Dergisi / 2020 / Cilt: 20 - Sayı: 4

Comparison of Machine Learning Methods in Prediction ofFinancial Failure of Businesses in The Manufacturing Industry:Evidence from Borsa İstanbul

İmalat Sanayi Sektöründeki İşletmelerin Finansal Başarısızlık Tahmininde Makine Öğrenmesi Yöntemlerinin Karşılaştırılması: Borsa İstanbul Örneği

Barış AKSOY
Derviş BOZTOSUN
Pages
237–268
Publication Date
2022-10-11
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)

The overall classification accuracy from the highest to the lowest of 3 years prior to successful-failure year (for 2006) is 84.21% for CART, 81.58% for ANN and 76.32% for C5.0, respectively.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, Artificial Neural Networks (NN), C5.0 Classification Algorithm, Classification and Regression Trees (CART) analyses were used to predict the financial success/failure of 126 businesses that are operating in the BIST (Borsa İstanbul) Manufacturing Industry Sector.
📊
Key Findings & Data:The data contains the years 2006 to 2009.
💡
Conclusion & Impact:The overall classification accuracy from the highest to the lowest of 3 years prior to successful-failure year (for 2006) is 84.21% for CART, 81.58% for ANN and 76.32% for C5.0, respectively.
🔬Design:Modeling & Simulation
📍Scope:2006 Verileri
⚙️Variables:Domain Parameters

Abstract

In this study, Artificial Neural Networks (NN), C5.0 Classification Algorithm, Classification and Regression Trees (CART) analyses were used to predict the financial success/failure of 126 businesses that are operating in the BIST (Borsa İstanbul) Manufacturing Industry Sector. The data contains the years 2006 to 2009. In the study, 25 quantitative variable and 4 qualitative variable were used. The overall classification accuracy from the highest to the lowest of 3 years prior to successful-failure year (for 2006) is 84.21% for CART, 81.58% for ANN and 76.32% for C5.0, respectively. The overall classification accuracy from the highest to the lowest of 2 years prior to successful-failure year (for 2007) is 86.84% for CART, 84.21% for ANN, 78.95% for C5.0, respectively. The overall classification accuracy from the highest to the lowest of 1 year prior to successful-failure year (for 2008) is 92.11% for CART, 92.11 for ANN and 86.84% for C5.0, respectively. ANN and CART models are notable in terms of their ability to predict upcoming financial failure of unsuccessful businesses with 100% classification accuracy from a year ago. The prediction of the financial success/failure by the three models obtained in the study more than one, two and three years ago shows that the models used in this study can be included in the model used by those concerned.

Özet

Bu çalışmada 2006-2009 yılları arasında BIST (Borsa İstanbul) İmalat Sanayi Sektöründe faaliyet gösteren 126işletmenin finansal başarı/başarısızlığını tahmin etmek üzere Yapay Sinir Ağları (NN), C5.0 Karar KuralıTüretme Algoritması, Sınıflama ve Regresyon Ağaçları (CART) analizi yapılmıştır. Çalışmada 25 mali tablolaradayalı nicel ve 4 mali tablolara dayalı olmayan nitel değişken kullanılmıştır. Analizler sonucunda başarılı- başarısızlık yılından 3 yıl öncesinde (2006 yılı için)en yüksekten düşüğe genel olarak (başarılı ve başarısıztoplamı) sırasıyla; CART %84.21, Yapay Sinir Ağları %81.58 ve C5.0 %76.32 sınıflandırma doğruluğunasahiptir. Başarılı-başarısızlık yılından 2 yıl öncesinde (2007 yılı için) en yüksekten düşüğe genel olarak sırasıyla;CART %86.84, Yapay Sinir Ağları %84.21 ve C5.0 %78.95 sınıflandırma doğruluğuna sahiptir. Başarılı- başarısızlık yılından 1 yıl öncesinde (2008 yılı için) en yüksekten düşüğe genel sınıflandırma doğruluğu sırasıyla;CART %92.11, Yapay Sinir Ağları %92.11 ve C5.0 %86.84 sınıflandırma doğruluğuna sahiptir. Çalışmada eldeedilen üç modelin finansal başarı/başarısızlığı bir, iki ve üç yıl öncesinden yüksek oranda tahmin etmesi ilgililerinkullandıkları modele bu çalışmada elde edilen modellerin de dâhil edilebileceğini göstermektedir.

🔖

Cite This Article

AKSOY, B., BOZTOSUN, D. (2022). Comparison of Machine Learning Methods in Prediction ofFinancial Failure of Businesses in The Manufacturing Industry:Evidence from Borsa İstanbul. Anadolu Üniversitesi Sosyal Bilimler Dergisi, Cilt: 20 - Sayı: 4, 237–268. https://ajindex.com/articles/439057

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (10)

  1. Akgüç, Ö. (1997).
  2. Financial Management, İstanbul:
  3. Avcıol publications. Aksoy, B. & Ertaş, F. C. (2016). Effects of corporate merger and acquisition on target company financial performance. Academic View Review, 54, 772-786.
  4. Aktaş, Ramazan, M. Mete Doğanay & Birol Yıldız (2003). Forecasting financial failure: comparison of statistical methods and artificial neural network. Ankara University SBF Journal, 58(4), 1-24. Access address: https://app.trdizin.gov.tr/publication/paper/detail/TWpZME1URT0
  5. Altman, E. I., T. Kant & T. Rattanaruengyot (2009). Bankruptcy performance: Avoiding”, Journal of Applied Corporate Finance, 21(3), 50-65. Access address: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1495296
  6. Baixauli, J. S. & Mo´dica-Milo, A. (2010). The bias of unhealthy smes ın bankruptcy prediction models. Journal of Small Business and Enterprise Development, 17(1), 60-77. Access address: https://www.emerald.com/insight/content/doi/10.1108/14626001011019134/full/pdf
  7. Beaver, William H. (1968). Market prices, financial ratios, and the prediction of failure. Journal of Accounting Research, 6(2), 179-192. Access address: https://www.jstor.org/stable/2490233?seq=1#metadata_info_tab_contents
  8. Bee, Thai Siew & Mehdi Abdollahi (2013). Corporate failure prediction: Malaysia’s emerging market. The International Journal of Finance, 25(4), 7985-8008. Access address: https://www.econbiz.de/Record/corporate-failure-prediction-malaysia-s-emerging-market-bee-thaisiew/ 10010382959 Bilir, H. (2015). Definition and market oriented solution of financial distress: Debt structuring, asset sales and new capital injection. Socioeconomic,1, 9-24. Access address: http://eds.a.ebscohost.com/eds/pdfviewer/pdfviewer?vid=0&sid=8f6841cb-c699-4e01-959df18901710107% 40sessionmgr4007 Brealey, R. & Myers, S. C., Trans. Bozkurt, A. J., Arıkan T., & Doğukanlı, H. (2007),
  9. Fundamentals of Business Finance, İstanbul: MC Graw Hill and Literature publications. Chandra, D. K., Ravi, V. & Bose, I. (2009). Failure prediction of DOTCOM companies using hybrid intelligent techniques. Expert Systems with Applications, 36, 4830-4837. Access address: https://www.sciencedirect.com/science/article/pii/S0957417408002741
  10. Chen, J., Marshall, B. R., Zhang, J., & Ganesh, S. (2006). Financial distress prediction in China. Review of Pacific Basin Financial Markets and Policies, 9(2), 317-336. Access address: https://econpapers.repec.org/article/wsirpbfmp/v_3a09_3ay_3a2006_3ai_3a02_3an_3as021909150600 0744.htm Chen,