BIST-30 Bankacılık Endeksi'nde Makine Öğrenmesi Kullanılarak Riske Maruz Değer Modelleri ile Piyasa Riski Analizi

Öz

Market riski bankaların ve portföy yöneticilerinin en dikkat ettiği risklerden birisidir. Basel kriterleri gereği Riske Maruz Değer (RMD) hesaplaması sık aralıklarla yapılmalıdır. RMD hesabı birkaç farklı yöntemle yapılabilirken yaklaşımlar ve modele eklenen değişkenler çok farklı olabilir. Makine öğrenmesi ve derin öğrenme yöntemlerindeki gelişmeler RMD hesaplamalarındaki çeşitliliği artırarak daha hassas ve karmaşık modellerin kurgulanmasına yardımcı olmuştur. Bu çalışmada BIST30’da yer alan 4 büyük bankanın (AKBNK, GARAN, ISCTR, YKBNK) hisselerinden Monte Carlo simülasyonu ve Random Forest yardımıyla portföy oluşturulmuştur. 5 yıllık ve günlük verilerle 10-ar günlük 126 periyot hesaplanmıştır. Son 4 periyot için 3 farklı RMD yöntemiyle (tarihi, parametrik ve Monte Carlo) tahminlemeler yapılmıştır. Değişken olarak VIX (korku endeksi), USD/TL, Gold/TL ve Brent/TL alınmıştır. Değişkenlerin uygunluğu makine öğrenmesi düzenlileştirme metotları olan Ridge, Lasso ve Elastic Net regresyon modelleriyle test edilmiştir. Bağımsız değişkenlerin hisselere etkisini ağırlıklandırılmasını ölçmek amacıyla yine Random Forest kullanılmıştır. Her bir RMD modeli için son 4 periyotta hisse ağırlık dağılımı çıkarılarak tekrar RMD sonuçları karşılaştırılmıştır. Bulgular sonucunda ilk periyot için parametrik RMD en iyi sonucu verirken diğer üç periyot için tarihi RMD en yakın sonucu vermiştir. Bulgular ile gerçekleşen sonuçlar kıyaslandığında bulguların daha iyimser olduğu, gerçekleşen değere en yakın sonuçların dahi %30’dan daha az oranda yaklaşmadığı gözlemlenmiştir. Farkın beklenenden fazla olmasının sebebi olarak banka hisselerinin son iki yılda değerinin altında olması ve seçilen son 4 dönemdeki borsa hareketlerinin -hisse özelinden bağımsız olarak- sert olması gösterilebilir.

Anahtar Kelimeler

• Piyasa Riski
• Riske Maruz Değer
• Makine Öğrenmesi
• Bist-30
• Bankacılık Endeksi

Market Risk Analysis with Value at Risk Models using Machine Learning in BIST-30 Banking Index

Öz

Market risk is one of the most critical risks for banks and portfolio managers. According to Basel criteria, Value at Risk (VaR) calculations should be conducted at regular intervals. VaR calculations can be performed using various methods, and the approaches and variables added to the model can vary significantly. Developments in machine learning and deep learning methods have increased the diversity of VaR calculations, enabling the construction of more accurate and complex models. In this study, a portfolio was created using the stocks of 4 major banks in BIST30 (AKBNK, GARAN, ISCTR, YKBNK) with the help of Monte Carlo simulation and Random Forest. Calculations were made for 126 periods with a 10-day interval using 5 years of daily data. Predictions were made for the last 4 periods using 3 different Value at Risk (VaR) methods (historical, parametric, and Monte Carlo). Independent variables such as VIX (fear index), USD/TL, Gold/TL, and Brent/TL were used. The suitability of the variables was tested with machine learning regularization methods, including Ridge, Lasso, and Elastic Net regression models. Random Forest was again used to measure the impact of independent variables on stocks' weights. For each VaR model, stock weight distributions were determined for the last 4 periods, and the realized VaR results were compared. As a result of the findings, the parametric VaR method provided the best result for the first period, while the historical VaR method provided the closest result for the other three periods. When comparing the findings with the actual results, it was observed that the findings were more optimistic, and even the closest results did not come within 30% of the actual value. The reason for the difference being greater than expected could be attributed to the fact that the value of bank stocks has been below their value in the last two years and the sharp movements in the stock market in the selected last 4 periods, independent of individual stocks.

Anahtar Kelimeler

• Market Risk
• Value at Risk
• Machine Learning
• BIST-30
• Banking Index

Etik Beyan

The study does not require an ethical statement.

Teşekkür

I would like to thank our graduate student Mehmet ÇELİK for his contributions.

Kaynakça

1. Acharya, V., Engle, R., & Richardson, M. (2012). Capital shortfall: A new approach to ranking and regulating systemic risks. American Economic Review, 102(3), 59-64. DOI: 10.1257/aer.102.3.59
2. Ahmed, L. (2015). The effect of foreign exchange exposure on the financial performance of commercial banks in Kenya. International journal of scientific and research publications, 5(11), 115-120.
3. Akan, N. B. (2007). Piyasa riski ölçümü. Bankacılar Dergisi, 61(59), 59-65.
4. Akhtaruzzaman, M., Boubaker, S., Chiah, M., & Zhong, A. (2021). COVID− 19 and oil price risk exposure. Finance research letters, 42, 101882. DOI: 10.1016/j.frl.2020.101882
5. Alexander, C. (2009). Market risk analysis Volume IV: Value-at-Risk models. Chichester: John Wiley & Sons.
6. Apergis, N., & Papoulakos, D. (2013). The Australian dollar and gold prices. The Open Economics Journal, 6(1). DOI: 10.2174/1874919401306010001
7. Apostolik, R., Donohue, C., & Went, P. (2009). Foundations of banking risk: an overview of banking, banking risks, and risk-based banking regulation. John Wiley.
8. Beutel, J., List, S., & von Schweinitz, G. (2019). Does machine learning help us predict banking crises? Journal of Financial Stability, 45, 100693. DOI: 10.1016/j.jfs.2019.100693

9. Cainelli, P. V., Pinto, A. C. F., & Klötzle, M. C. (2020). Study on the relationship between the IVol-BR and the future returns of the Brazilian stock market. Revista Contabilidade & Finanças, 32, 255-272. DOI: 10.1590/1808-057x202009890
10. Carmo, B. B. T. D., Medeiros, P. P. M. D., Gonçalo, T. E. E., & Correia, G. F.. (2023, June 12). Framework to assist investment portfolio generation for financial sector. Exacta, 21(2), 337-365. DOI: 10.5585/exactaep.2021.18687
11. Chakraborty, G., Chandrashekhar, G. R., & Balasubramanian, G. (2021). Measurement of extreme market risk: Insights from a comprehensive literature review. Cogent Economics & Finance, 9(1), DOI: 10.1080/23322039.2021.1920150
12. Chatzis, S. P., Siakoulis, V., Petropoulos, A., Stavroulakis, E., & Vlachogiannakis, N. (2018). Forecasting stock market crisis events using deep and statistical machine learning techniques. Expert systems with applications, 112, 353-371. DOI: 10.1016/j.eswa.2018.06.032
13. Dağlı, Hüseyin (2004), Sermaye Piyasası ve Portföy Analizi, 2. Baskı, Derya Kitabevi, Trabzon. Daniali, S. M., Barykin, S. E., Kapustina, I. V., Mohammadbeigi Khortabi, F., Sergeev, S. M., Kalinina, O. V., & Senjyu, T. (2021). Predicting volatility index according to technical index and economic indicators on the basis of deep learning algorithm. Sustainability, 13(24), 14011. DOI: 10.3390/su132414011
14. Deng, S., Mitsubuchi, T., & Sakurai, A. (2014). Stock price change rate prediction by utilizing social network activities. The Scientific World Journal, 2014. DOI: 10.1155/2014/861641
15. Döpke, J., Fritsche, U., & Pierdzioch, C. (2017). Predicting recessions with boosted regression trees. International Journal of Forecasting, 33(4), 745-759. DOI: 10.1016/j.ijforecast.2017.02.003
16. Foroni, B., Morelli, G., & Petrella, L. (2022). The network of commodity risk. Energy Systems, 1-47. DOI: 10.1007/s12667-022-00530-7
17. Glasserman, P. (2004) Monte Carlo Methods in Financial Engineering. Springer-Verlag, New York
18. Groth, S. S., & Muntermann, J. (2011). An intraday market risk management approach based on textual analysis. Decision support systems, 50(4), 680-691. DOI: 10.1016/j.dss.2010.08.019
19. Höçük, F. (2022). Incorporation of Foreign Exchange Risk to Fama-French Factor Model: A Study on Borsa İstanbul (Master's thesis, Middle East Technical University).
20. Hu, Z., Zhao, Y., & Khushi, M. (2021). A survey of forex and stock price prediction using deep learning. Applied System Innovation, 4(1), 9. DOI: 10.3390/asi4010009
21. İskenderoglu, Ö., & Akdağ, S. (2020). Comparison of the effect of VIX fear index on stock exchange indices of developed and developing countries: The G20 case. The South East European Journal of Economics and Business, 15(1), 105-121. DOI: 10.2478/jeb-2020-0009
22. Jorion, P. (1996). Risk2: Measuring the risk in value at risk. Financial analysts journal, 52(6), 47-56. DOI: 10.2469/faj.v52.n6.2039
23. Köseoğlu B. (2020, Oct 26). Ridge, Lasso ve Elastic Net https://buse-koseoglu13.medium.com/ridge-lasso-ve-elastic-net-b6089bf2f09 Access Date: 31.10.2023
24. Leo, M., Sharma, S., & Maddulety, K. (2019). Machine learning in banking risk management: A literature review. Risks, 7(1), 29. DOI: 10.3390/risks7010029
25. Lessmann, S., Baesens, B., Seow, H. V., & Thomas, L. C. (2015). Benchmarking state-of-the-art classification algorithms for credit scoring: An update of research. European Journal of Operational Research, 247(1), 124-136. DOI: 10.1016/j.ejor.2015.05.030
26. Lu, C., Teng, Z., Gao, Y., Wu, R., Hossain, M. A., & Fang, Y. (2022). Analysis of early warning of RMB exchange rate fluctuation and value at risk measurement based on deep learning. Computational Economics, 59(4), 1501-1524. DOI: 10.1007/s10614-021-10172-z
27. Markowitz, H. (1952) Portfolio Selection. The Journal of Finance, Vol. 7, No. 1, pp. 77-91. March. 1952.
28. Narasimhan, M., & Viswanathan, S. (2011). The VIX as a forecasting tool. Journal of Futures Markets, 31(1), 23-48. Ni, L., Li, Y., Wang, X., Zhang, J., Yu, J., & Qi, C. (2019). Forecasting of forex time series data based on deep learning. Procedia computer science, 147, 647-652. DOI: 10.1016/j.procs.2019.01.189
29. Noorian, F., & Leong, P. H. (2014, March). Dynamic hedging of foreign exchange risk using stochastic model predictive control. In 2014 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) (pp. 441-448). IEEE.
30. Öcal, U. & Demireli, E. (2010). Risk Bileşenleri Analizi: İmkb’de Bir Uygulama. Uluslararası Yönetim İktisat ve İşletme Dergisi, 6(12), 25-36.
31. Poitras, G. (2013). Commodity risk management: Theory and application. Routledge.
32. Rupeika-Apoga, R. (2005). Nowadays Approach to Foreign Exchange Risk Management. Management of Organizations: Systematic Research, (35).
33. Russon, M. G., & Vakil, A. F. (2017). On The Non-Linear Relationship Between VIX And Realized Sp500 Volatility. Invest. Manag. Financ. Innov, 14, 200-206.
34. Sarangi, P. K., Chawla, M., Ghosh, P., Singh, S., & Singh, P. K. (2022). FOREX trend analysis using machine learning techniques: INR vs USD currency exchange rate using ANN-GA hybrid approach. Materials Today: Proceedings, 49, 3170-3176. DOI: 10.1016/j.matpr.2020.10.960
35. Shaikh, I., & Padhi, P. (2015). The implied volatility index: Is ‘investor fear gauge’or ‘forward-looking’?. Borsa Istanbul Review, 15(1), 44-52. DOI: 10.1016/j.bir.2014.10.001
36. Simonella, R., & Vázquez, C. (2023). XVA in a multi-currency setting with stochastic foreign exchange rates. Mathematics and Computers in Simulation, 207, 59-79. DOI: 10.1016/j.matcom.2022.12.014
37. Su, S. (2018, June). An investigation of foreign exchange risk management in Chinese multinational companies compared with US and UK MNEs. In 2018 2nd International Conference on Management, Education and Social Science (ICMESS 2018) (pp. 530-534). Atlantis Press.
38. Whaley, R. E. (2000). The investor fear gauge. Journal of portfolio management, 26(3), 12.
39. Widhiarti, R. P., Anggraeni, L., & Pasaribu, S. H. (2018). Analysis of investor sentiment impact in Indonesia composite stock price index return volatility. Indonesian Journal of Business and Entrepreneurship (IJBE), 4(3), 239-239. DOI: 10.17358/ijbe.4.3.239
40. Zhang, K., Wang, X., Wang, J., Wang, S., & Hui, F. (2022). Analysis and Prediction of Corporate Finance and Exchange Rate Correlation Based on Machine Learning Algorithms. Computational Intelligence and Neuroscience, 2022. DOI: 10.1155/2022/2850604
41. Zhang, Y., & Hamori, S. (2020). The predictability of the exchange rate when combining machine learning and fundamental models. Journal of Risk and Financial Management, 13(3), 48. DOI: 10.3390/jrfm13030048