Gönüllü Karbon Açıklaması Tahmininde Finansal Göstergelerin Rolü: Makine Öğrenmesi Yöntemleri ile Karşılaştırmalı Bir Analiz

Öz

Sanayi Devriminden bu yana atmosferdeki karbondioksit emisyonları ve ormansızlaşmanın iklim değişikliğinin başlıca nedenleri olduğu düşünülmektedir. Birçok ülke sera gazı emisyonlarını azaltmak için politikalar geliştirmekte ve firmaları karbon emisyonlarını açıklamaya ve azaltmaya teşvik etmektedir. Bu çalışma, makine öğrenmesi yöntemlerini kullanarak, 2016-2023 yılları arasında Borsa İstanbul'da işlem gören firmaların CDP (Carbon Disclosure Project) anketine yanıt verme istekliliği ile ölçülen karbon riski farkındalığının potansiyel finansal belirleyicilerini ortaya çıkarmayı amaçlamaktadır. Çalışmanın bulguları, firmaların finansal göstergelerine dayalı modeller aracılığıyla, doğrusal olmayan, topluluk öğrenmesi tabanlı Rastgele Orman ve XGBoost algoritmaları kullanılarak, gönüllü karbon açıklaması yapma eğilimlerinin %92’nin üzerinde bir doğruluk oranıyla tahmin edilebildiğini ortaya koymaktadır. Ayrıca açıklanabilir yapay zekâ araçları kullanılarak yapılan analizler, özkaynakların toplam borçlara oranı, duran varlıkların özkaynaklara oranı ve uzun vadeli borçların toplam borçlara oranı gibi belirli finansal oranların, XGBoost algoritmasında modelin açıklayıcılığına önemli düzeyde katkı sağladığını göstermektedir. Son olarak, çalışma, makine öğrenmesi algoritmalarının kurumsal karbon emisyonlarının tahmininde yatırımcıların risk analizini iyileştirme potansiyeline ve bu bulgunun sürdürülebilir yatırım stratejilerinin hem kuramsal hem de uygulamalı olarak geliştirilmesine katkı sunabileceğine dikkat çekmektedir.

Anahtar Kelimeler

• İklim Değişikliği
• Gönüllü Karbon Saydamlığı
• Makine Öğrenmesi
• Açıklanabilir Yapay Zekâ
• Borsa İstanbul

The Role of Financial Indicators in the Prediction of Voluntary Carbon Disclosure: A Comparative Analysis with Machine Learning Methods

Öz

Since the Industrial Revolution, carbon dioxide emissions and deforestation have been considered the primary causes of climate change. Many countries are developing policies to reduce greenhouse gas emissions and are encouraging firms to disclose and reduce their carbon emissions. This study aims to identify the potential financial determinants of carbon risk awareness, as measured by the willingness to respond to the CDP (Carbon Disclosure Project) survey, among firms listed on the Borsa Istanbul between 2016 and 2023, using machine learning methods. The findings reveal that whether firms will make voluntary carbon disclosures can be predicted with an accuracy rate exceeding 92% using nonlinear, ensemble learning-based Random Forest and XGBoost algorithms in models based on financial indicators. Furthermore, analyses conducted with explainable artificial intelligence tools indicate that specific financial ratios, such as the ratio of equity to total debt, the ratio of fixed assets to equity, and the ratio of long-term debt to total debt, significantly enhance the model's explainability within the XGBoost algorithm. Finally, the study highlights the potential of machine learning algorithms to improve investors' risk analysis in predicting corporate carbon emissions and demonstrates that this finding contributes to both the theoretical and practical development of sustainable investment strategies.

Anahtar Kelimeler

• Climate Change
• Voluntary Carbon Disclosure
• Machine Learning
• Explainable Artificial Intelligence
• Borsa İstanbul

Kaynakça

1. Alsaifi, K., Elnahass, M. and Salama, A. (2020). Carbon disclosure and financial performance: UK environmental policy. Business Strategy and the Environment, 29(2), 711-726. https://doi.org/10.1002/bse.2426
2. Anquetin, T., Coqueret, G., Tavin, B. and Welgryn, L. (2022). Scopes of carbon emissions and their impact on green portfolios. Economic Modelling, 115, 105951. https://doi.org/10.1016/j.econmod.2022.105951
3. Arendt, C.A., Hyland, E.G. and Piliouras, A. (2021). The geological consequences of global climate change. In D. Alderton and S.A. Elias (Eds.), Encyclopedia of geology second edition (pp. 510–522). New York: Academic Press.
4. Balbal, K.F. (2024). MIPART: A partial decision tree-based method for multiple-instance classification. Applied Sciences, 14(24), 11696. https://doi.org/10.3390/app142411696
5. Benkraiem, R., Shuwaikh, F., Lakhal, F. and Guizani, A. (2022). Carbon performance and firm value of the World's most sustainable companies. Economic Modelling, 116, 106002. https://doi.org/10.1016/j.econmod.2022.106002
6. Birkey, R.N., Michelon, G., Patten, D.M. and Sankara, J. (2016). Does assurance on CSR reporting enhance environmental reputation? An examination in the U.S. context. Accounting Forum, 40(3), 143–152. https://doi.org/10.1016/j.accfor.2016.07.001
7. Bolton, P. and Kacperczyk, M. (2021). Do investors care about carbon risk? Journal of Financial Economics, 142(2), 517-549. https://doi.org/10.1016/j.jfineco.2021.05.008
8. Bose, S., Minnick, K. and Shams, S. (2021). Does carbon risk matter for corporate acquisition decisions? Journal of Corporate Finance, 70, 102058. https://doi.org/10.1016/j.jcorpfin.2021.102058

9. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/A:1010933404324
10. Caby, J., Ziane, Y. and Lamarque, E. (2020). The determinants of voluntary climate change disclosure commitment and quality in the banking industry. Technological Forecasting and Social Change, 161, 120282. https://doi.org/10.1016/j.techfore.2020.120282
11. Chen, T. and Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In B. Krishnapuram and M. Shah (Eds.), Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 785-794). New York: Association for Computing Machinery.
12. Cho, C.H., Guidry, R.P., Hageman, A.M. and Patten, D.M. (2012). Do actions speak louder than words? An empirical investigation of corporate environmental reputation. Accounting, Organizations and Society, 37(1), 14-25. https://doi.org/10.1016/j.aos.2011.12.001
13. Choi, B.B., Lee, D. and Psaros, J. (2013). An analysis of Australian company carbon emission disclosures. Pacific Accounting Review, 25(1), 58-79. https://doi.org/10.1108/01140581311318968
14. Cooper, S.A., Raman, K.K. and Yin, J. (2018). Halo effect or fallen angel effect? Firm value consequences of greenhouse gas emissions and reputation for corporate social responsibility. Journal of Accounting and Public Policy, 37(3), 226-240. https://doi.org/10.1016/j.jaccpubpol.2018.04.003
15. Costa de Oliveira, A., Marini, N. and Farias, D.R. (2014). Climate change: New breeding pressures and goals. In N.K. Van Alfen (Ed.), Encyclopedia of agriculture and food systems (pp. 284–293). New York: Academic Press.
16. Cristianini, N. and Ricci, E. (2008). Support vector machines. In M.-Y. Kao (Ed.), Encyclopedia of algorithms (pp. 928–932). US: Springer.
17. D’Amato, V., D’Ecclesia, R. and Levantesi, S. (2021). Fundamental ratios as predictors of ESG scores: A machine learning approach. Decisions in Economics and Finance, 44(2), 1087-1110. https://doi.org/10.1007/s10203-021-00364-5
18. Deisenroth, M.P., Faisal, A.A. and Ong, C.S. (2020). Mathematics for machine learning. Cambridge: Cambridge University Press.
19. Desai, R., Raval, A., Baser, N. and Desai, J. (2022). Impact of carbon emission on financial performance: Empirical evidence from India. South Asian Journal of Business Studies, 11(4), 450-470. https://doi.org/10.1108/SAJBS-10-2020-0384
20. Dhanalakshmi, P., Palanivel, S. and Ramalingam, V. (2009). Classification of audio signals using SVM and RBFNN. Expert Systems with Applications, 36(3), 6069-6075. https://doi.org/10.1016/j.eswa.2008.06.126
21. Fan, P., Qian, X. and Wang, J. (2023). Does gender diversity matter? Female directors and firm carbon emissions in Japan. Pacific-Basin Finance Journal, 77, 101931. https://doi.org/10.1016/j.pacfin.2022.101931
22. Frank, E., Trigg, L., Holmes, G. and Witten, I.H. (2000). Naive Bayes for regression. Machine Learning, 41, 5-25. https://doi.org/10.1023/A:1007670802811
23. Frost, G., Jones, S. and Yu, M. (2023). Voluntary carbon reporting prediction: A machine learning approach. Abacus, 59(4), 1116-1166. https://doi.org/10.1111/abac.12298
24. Füssel, H.M. and Klein, R.J. (2006). Climate change vulnerability assessments: An evolution of conceptual thinking. Climatic Change, 75(3), 301-329. https://doi.org/10.1007/s10584-006-0329-3
25. Griffin, P.A., Lont, D.H. and Sun, E.Y. (2017). The relevance to investors of greenhouse gas emission disclosures. Contemporary Accounting Research, 34(2), 1265-1297. https://doi.org/10.1111/1911- 3846.12298
26. Griffin, P.A., Neururer, T. and Sun, E.Y. (2020). Environmental performance and analyst information processing costs. Journal of Corporate Finance, 61, 101397. https://doi.org/10.1016/j.jcorpfin.2018.08.008
27. Guastella, G., Mazzarano, M., Pareglio, S. and Xepapadeas, A. (2022). Climate reputation risk and abnormal returns in the stock markets: A focus on large emitters. International Review of Financial Analysis, 84, 102365. https://doi.org/10.1016/j.irfa.2022.102365
28. Güneysu, Y. ve Atasel, O.Y. (2022). Karbon emisyonları ile finansal performans arasındaki ilişkinin incelenmesi: BIST100 endeksinde bir araştırma. Fırat Üniversitesi Sosyal Bilimler Dergisi, 32, 3(1183-1193). https://doi.org/10.18069/firatsbed.1125859
29. Haque, F. (2017). The effects of board characteristics and sustainable compensation policy on carbon performance of UK firms. The British Accounting Review, 49(3), 347-364. https://doi.org/10.1016/j.bar.2017.01.001
30. Harker, C., Hassall, M., Lant, P., Rybak, N. and Dargusch, P. (2022). What can machine learning teach us about Australian climate risk disclosures? Sustainability, 14(16), 10000. https://doi.org/10.3390/su141610000
31. Harrison, M. (2024). Effective XGBoost: Optimizing, tuning, understanding, and deploying classification models. Salt Lake City: MetaSnake.
32. History of the convention. (n.d.). United Nations framework convention on climate change. Retrieved from https://unfccc.int/process/the-convention/history-of-the-convention#Climate-Change-in-context
33. Hollindale, J., Kent, P., Routledge, J. and Chapple, L. (2019). Women on boards and greenhouse gas emission disclosures. Accounting & Finance, 59(1), 277-308. https://doi.org/10.1111/acfi.12258
34. Hossin, M. and Sulaiman, M.N. (2015). A review on evaluation metrics for data classification evaluations. International Journal of Data Mining & Knowledge Management Process, 5(2), 1-11. https://doi.org/10.5121/ijdkp.2015.5201
35. Huang, X., Jin, G. and Ruan, W. (2023). Naive Bayes. In X. Huang, G. Jin and W. Ruan (Eds.), Machine learning safety. Artificial intelligence: Foundations, theory, and algorithms (pp. 95–102). Berlin: Springer Nature.
36. Jung, J., Herbohn, K. and Clarkson, P. (2018). Carbon risk, carbon risk awareness and the cost of debt financing. Journal of Business Ethics, 150, 1151-1171. https://doi.org/10.1007/s10551-016-3207-6
37. Kabir, M.N., Rahman, S., Rahman, M.A. and Anwar, M. (2021). Carbon emissions and default risk: International evidence from firm-level data. Economic Modelling, 103, 105617. https://doi.org/10.1016/j.econmod.2021.105617
38. Kleimeier, S. and Viehs, M. (2021). Pricing carbon risk: Investor preferences or risk mitigation? Economics Letters, 205, 109936. https://doi.org/10.1016/j.econlet.2021.109936
39. Kolk, A. and Pinkse, J. (2005). Business responses to climate change: Identifying emergent strategies. California Management Review, 47(3), 6-20. https://doi.org/10.2307/41166304
40. Krishnamurti, C. and Velayutham, E. (2018). The influence of board committee structures on voluntary disclosure of greenhouse gas emissions: Australian evidence. Pacific-Basin Finance Journal, 50, 65-81. https://doi.org/10.1016/j.pacfin.2017.09.003
41. Kyoto Protocol. (n.d.). Climate change presidency. Retrieved from https://iklim.gov.tr/kyoto-protokolu-i-35#
42. Lee, K.H., Min, B. and Yook, K.H. (2015). The impacts of carbon (CO2) emissions and environmental research and development (R&D) investment on firm performance. International Journal of Production Economics, 167, 1-11. https://doi.org/10.1016/j.ijpe.2015.05.018
43. Liao, L., Luo, L. and Tang, Q. (2015). Gender diversity, board independence, environmental committee and greenhouse gas disclosure. The British Accounting Review, 47(4), 409-424. https://doi.org/10.1016/j.bar.2014.01.002
44. Loh, W.Y. (2011). Classification and regression trees. WIREs Data Mining and Knowledge Discovery, 1(1), 14-23. https://doi.org/10.1002/widm.8
45. López-Pacheco, I.Y., Rodas-Zuluaga, L.I., Fuentes-Tristan, S., Castillo-Zacarías, C., Sosa-Hernández, J.E., Barceló, D., … Parra-Saldívar, R. (2021). Phycocapture of CO2 as an option to reduce greenhouse gases in cities: Carbon sinks in urban spaces. Journal of CO2 Utilization, 53, 101704. https://doi.org/10.1016/j.jcou.2021.101704
46. Lu, W., Zhu, N. and Zhang, J. (2021). The impact of carbon disclosure on financial performance under low carbon constraints. Energies, 14 (14). https://doi.org/10.3390/en14144126
47. Matsumura, E.M., Prakash, R. and Vera-Munoz, S.C. (2014). Firm-value effects of carbon emissions and carbon disclosures. The Accounting Review, 89(2), 695-724. https://doi.org/10.2308/accr-50629
48. Merello, P., Barberá, A. and De la Poza, E. (2022). Is the sustainability profile of FinTech companies a key driver of their value? Technological Forecasting and Social Change, 174, 121290. https://doi.org/10.1016/j.techfore.2021.121290
49. Müller, A.C. and Guido, S. (2016). Introduction to machine learning with Python: A guide for data scientists (1st ed.). California: O'Reilly Media.
50. NASA. (2024). What is climate change? Retrieved from https://science.nasa.gov/climate-change/what-is-climate-change/
51. Nguyen, J.H. and Phan, H.V. (2020). Carbon risk and corporate capital structure. Journal of Corporate Finance, 64, 101713. https://doi.org/10.1016/j.jcorpfin.2020.101713
52. Paris Anlaşması. (n.d.). T.C. Dışişleri Bakanlığı. Retrieved from https://www.mfa.gov.tr/paris-anlasmasi.tr.mfa
53. Pires, J.C.M. (2017). COP21: The algae opportunity? Renewable and Sustainable Energy Reviews, 79, 867–877. https://doi.org/10.1016/J.RSER.2017.05.197
54. Rahat, B. and Nguyen, P. (2022). Risk-adjusted investment performance of green and black portfolios and impact of toxic divestments in emerging markets. Energy Economics, 116, 106423. https://doi.org/10.1016/j.eneco.2022.106423
55. Rainio, O., Teuho, J. and Klén, R. (2024). Evaluation metrics and statistical tests for machine learning. Scientific Reports, 14(1), 6086. https://doi.org/10.1038/s41598-024-56706-x
56. Rogoff, M.J. (2014). Introduction. In M.J. Rogoff (Ed.), Solid waste recycling and processing (pp. 1–9). Amsterdam: William Andrew.
57. Rohleder, M., Wilkens, M. and Zink, J. (2022). The effects of mutual fund decarbonization on stock prices and carbon emissions. Journal of Banking & Finance, 134, 106352. https://doi.org/10.1016/j.jbankfin.2021.106352
58. Safiullah, M., Kabir, M.N. and Miah, M.D. (2021). Carbon emissions and credit ratings. Energy Economics, 100, 105330. https://doi.org/10.1016/j.eneco.2021.105330
59. Sagi, O. and Rokach, L. (2021). Approximating XGBoost with an interpretable decision tree. Information Sciences, 572, 522-542. https://doi.org/10.1016/j.ins.2021.05.055
60. Sangiorgi, I. and Schopohl, L. (2021). Why do institutional investors buy green bonds: Evidence from a survey of European asset managers. International Review of Financial Analysis, 75, 101738. https://doi.org/10.1016/j.irfa.2021.101738
61. Sarang, P. (2023). Naive Bayes. In O. Marques (Ed.), Thinking data science: A data science practitioner's guide (pp. 143–152). Cham: Springer International Publishing.
62. The Paris Agreement. (n.d.). UNFCCC. Retrieved from https://unfccc.int/process-and-meetings/the-paris-agreement
63. Vapnik, V. (1998). Statistical learning theory. New York: Wiley.
64. Velte, P., Stawinoga, M. and Lueg, R. (2020). Carbon performance and disclosure: A systematic review of governance-related determinants and financial consequences. Journal of Cleaner Production, 254, 120063. https://doi.org/10.1016/j.jclepro.2020.120063
65. Yu, B., Li, C., Mirza, N. and Umar, M. (2021). Forecasting credit ratings of decarbonized firms: Comparative assessment of machine learning models. Technological Forecasting and Social Change, 174, 121255. https://doi.org/10.1016/j.techfore.2021.121255
66. Zhang, H. (2004). The optimality of Naive Bayes. Paper presented at the Seventeenth international Florida artificial intelligence research society conference (FLAIRS 2004). Menlo Park, California. Retrieved from https://cdn.aaai.org/FLAIRS/2004/Flairs04-097.pdf