摘要(英) |
In recent years, due to global climate and social changes, in addition to traditional accounting indicators, Environmental, Social, Governance (ESG) has also become an important indicator for evaluating enterprise development. ESG evaluates the sustainable development of a company from the perspective of sustainable development. Environmental mainly examines the company′s assessment of issues related to the natural environment, including efforts to address issues such as green energy, exhaust emissions, and environmental pollution. Social mainly examines the company′s assessment of social-related responsibilities, including management results on issues such as employees, products, and supply chains. Governance mainly reviews the company′s evaluation of internal management-related issues, including the board of directors, shareholders, corporate ethics and other related issues. Many studies have found that ESG Rating is positively correlated with the stock price of companies, and companies with higher ESG Ratings have lower investment risks. Therefore, many investors used ESG Rating as the basis for their investment decisions. It can be seen that the ESG Rating of a company will have a very significant impact on its development.
As the technology of natural language processing becomes more and more mature, its related applications become more diverse. This research aims at the ESG Rating assessed by Morgan Stanley Capital International (MSCI). This study will use BERT natural language processing technology to conduct ESG analysis on the company′s public reports, so as to understand the correlation between the company and each Key Issue defined by MSCI. Finally, try to predict the MSCI ESG Rating of enterprises through machine learning and deep learning model methods. |
參考文獻 |
Ashwin Kumar, N. C., Smith, C., Badis, L., Wang, N., Ambrosy, P., & Tavares, R. (2016). ESG factors and risk-adjusted performance: A new quantitative model. Journal of Sustainable Finance & Investment, 6(4), 292–300.
Avetisyan, E., & Hockerts, K. (2017). The consolidation of the ESG rating industry as an enactment of institutional retrogression. Business Strategy and the Environment, 26(3), 316–330.
Baier, P., Berninger, M., & Kiesel, F. (2020). Environmental, social and governance reporting in annual reports: A textual analysis. Financial Markets, Institutions & Instruments, 29(3), 93–118.
Beltagy, I., Lo, K., & Cohan, A. (2019). SciBERT: A pretrained language model for scientific text. ArXiv Preprint ArXiv:1903.10676.
Consolandi, C., Eccles, R. G., & Gabbi, G. (2020). How material is a material issue? Stock returns and the financial relevance and financial intensity of ESG materiality. Journal of Sustainable Finance & Investment, 1–24.
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. ArXiv Preprint ArXiv:1810.04805.
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020). BioBERT: A pre-trained biomedical language representation model for biomedical text mining. Bioinformatics, 36(4), 1234–1240.
Maiti, M. (2021). Is ESG the succeeding risk factor? Journal of Sustainable Finance & Investment, 11(3), 199–213.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Lukasz, & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.
Yang, Z., Dai, Z., Yang, Y., Carbonell, J., Salakhutdinov, R. R., & Le, Q. V. (2019). Xlnet: Generalized autoregressive pretraining for language understanding. Advances in Neural Information Processing Systems, 32. |