學術產出-Theses

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

題名 運用機器學習模型分析影響公司風險的ESG因子:以台灣市場為例
Machine Learning application on the ESG factor analysis on Firm Risks
作者 孫嘉蔚
Sun, Chia-Wei
貢獻者 楊曉文
Yang, Sharon. S.
孫嘉蔚
Sun, Chia-Wei
關鍵詞 機器學習
ESG
公司風險
股價崩跌風險
Machine Learning
ESG
Firm Risk
Stock Crash Risk
日期 2021
上傳時間 1-Oct-2021 10:03:16 (UTC+8)
摘要 本研究採用機器學習模型,以模型找出何種ESG因子對於台灣企業的公司風險與股價崩跌有較高的解釋力,拆分企業社會責任對於風險的影響。本次研究用梯度提升決策樹(Gradient Boost Trees,以下稱GDBT)、XGBoost、隨機森林(Random Forest),並使用Refinitiv資料庫中的綜合分數、環境分數、社會分數與公司治理分數下的14個指標作為ESG變數。我採用公司特有風險與兩項股價崩跌風險指標(NCSKEW、DUVOL)作為風險變數,以2010-2019年間的992筆台灣公司股價資料計算而成,期望能探究機器學習模型在ESG評分與公司個別風險的效果。
而實證結果顯示,使用XGBoost模型與GDBT在對公司風險與股價崩跌風險模型解釋力上比起隨機森林有較佳的表現。進一步透過分析因子重要性後,數據結果顯示ESG分數綜合指標如ESG綜合分數在公司風險的重要度表現較不佳,顯示比起採用社會責任細項指標,投資人若想依照綜合分數作為投資組合風險管理考量,較無效率。社會類別如企業社會責任策略分數、社區分數與人權分數因子中,在公司特有風險與股價崩跌風險當中,皆具有一定程度的影響性,可以作為企業內部風險管理考量上的指標依據。
This study approaches three machine learning models to find out which ESG perfor-mance factors have significant impact on firm specific risk and stock crash risks ( NCSKEW,DUVOL ). Three models such as Gradient Boost Trees (Gradient Boost Trees, hereinafter referred to as GDBT), XGBoost, and Random Forest are applied into analyzing the effect of 14 ESG Performance from Reuters database on firm specific risks and stock crash risks. The sample of the study is mainly based Taiwanese firms in Refinitiv ESG database, ranging from the period of 2010 to 2019.The empirical results show that XGBoost and GDBT have the better performance than Random Forest in ex-plaining the company risk and stock crash risk. Through factor importance analysis, I found that combined ESG score are less important in the part of firm risks. This shows that rather than taking ESG composite score into account, investors should consider in-dividual dimensions of Environmental, Social, and Governance indicators for further portfolio risk management.
Social categories, such as corporate social responsibility strategy scores, communi-ty scores, and human rights score, have a certain degree of influence in the firm specific risks and stock crash risks. These scores could be indicators for internal risk manage-ment on the scope of portfolio management and firm risk management.
參考文獻 郭怡萱(2018)。ESG績效對公司風險之影響。國立政治大學財務管理所碩士論文,台北市。
國立臺北大學. (2021). 台灣永續投資調查. http://www.aacsb.ntpu.edu.tw/twsvi/uploads/file/cus2_zu55548c6z.pdf
Annisa, A. N., & Hartanti, D. (2021). The Impact of Environmental, Social, and Gov-ernance Performance on Firm Risk in the ASEAN-5 Countries, 2011-2017. In Asia-Pacific Research in Social Sciences and Humanities Universitas Indonesia Conference (APRISH 2019) (pp. 625-634). Atlantis Press.
Antoncic, M., Bekaert, G., Rothenberg, R. V., & Noguer, M. (2020). Sustainable In-vestment-Exploring the Linkage between Alpha, ESG, and SDG`s. ESG, and SDG`s (August 2020).
Ballings, M., Van den Poel, D., Hespeels, N., & Gryp, R. (2015). Evaluating multiple classifiers for stock price direction prediction. Expert systems with applications, 42(20), 7046-7056.
Broadstock, D. C., Chan, K., Cheng, L. T., & Wang, X. (2021). The role of ESG per-formance during times of financial crisis: Evidence from COVID-19 in China. Fi-nance research letters, 38, 101716.
Chen, J., Hong, H., & Stein, J. C. (2001). Forecasting crashes: Trading volume, past returns, and conditional skewness in stock prices. Journal of financial economics, 61(3), 345-381.
Chen, T., & Guestrin, C. (2016, August). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).
Chen, Q., & Liu, X. Y. (2020, May). Quantifying ESG Alpha in Scholar Big Data: An Automated Machine Learning Approach. In ACM International Conference on AI in Finance.
Crane, A., Matten, D., & Moon, J. (2008). Corporations and citizenship: Business, re-sponsibility and society. Cambridge University Press.
De Lucia, C., Pazienza, P., & Bartlett, M. (2020). Does good ESG lead to better finan-cial performances by firms? Machine learning and logistic regression models of public enterprises in Europe. Sustainability, 12(13), 5317.
Edmans, A. (2011). Does the stock market fully value intangibles? Employee satisfac-tion and equity prices. Journal of Financial economics, 101(3), 621-640.
Ferriani, F., & Natoli, F. (2020). ESG risks in times of COVID-19. Applied Econom-ics Letters, 1-5.
Friedman, J. H. (2001). Greedy function approximation: a gradient boosting ma-chine. Annals of statistics, 1189-1232.
Godfrey, P. C. (2005). The relationship between corporate philanthropy and sharehold-er wealth: A risk management perspective. Academy of management review, 30(4), 777-798.
Godfrey, P. C., Merrill, C. B., & Hansen, J. M. (2009). The relationship between cor-porate social responsibility and shareholder value: An empirical test of the risk management hypothesis. Strategic management journal, 30(4), 425-445.
Guo, T. (2020). Esg2risk: A deep learning framework from esg news to stock volatility prediction. Available at SSRN 3593885.
Harvey, C. R., & Siddique, A. (2000). Conditional skewness in asset pricing tests. The Journal of finance, 55(3), 1263-1295.
Ho, T. K. (1995, August). Random decision forests. In Proceedings of 3rd international conference on document analysis and recognition (Vol. 1, pp. 278-282). IEEE.
Huseynov, F., & Klamm, B. K. (2012). Tax avoidance, tax management and corporate social responsibility. Journal of Corporate Finance, 18(4), 804-827.
Ilhan, E., Sautner, Z., & Vilkov, G. (2021). Carbon tail risk. The Review of Financial Studies, 34(3), 1540-1571.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Jo, H., & Na, H. (2012). Does CSR reduce firm risk? Evidence from controversial in-dustry sectors. Journal of Business Ethics, 110(4), 441-456.
Khaidem, L., Saha, S., & Dey, S. R. (2016). Predicting the direction of stock market prices using random forest. arXiv preprint arXiv:1605.00003.
Kim, Y., Li, H., & Li, S. (2014). Corporate social responsibility and stock price crash risk. Journal of Banking & Finance, 43, 1-13.
KPMG. (2020). The KPMG Survey of Sustainability Reporting 2020. https://assets.kpmg/content/dam/kpmg/xx/pdf/2020/11/the-time-has-come.pdf
Liu, H. (2020). Stock Selection Strategy Based on Support Vector Machine and eX-treme Gradient Boosting Methods. 2020 the 4th International Conference on Big Data Research, 36-39.
Margot, V., Geissler, C., de Franco, C., & Monnier, B. (2021). ESG Investments: Fil-tering versus Machine Learning Approaches. Applied Economics and Finance, 8(2), 1-16.
Minor, D., & Morgan, J. (2011). CSR as reputation insurance: Primum non nocere. California Management Review, 53(3), 40-59.
Mitsuzuka, K., Ling, F., & Ohwada, H. (2017, February). Analysis of CSR activities affecting corporate value using machine learning. In Proceedings of the 9th Inter-national Conference on Machine Learning and Computing (pp. 11-14).
Murata, Hamori, 2021. ESG Disclosures and Stock Price Crash Risk. Journal of Risk and Financial Management, vol. 14(2), pages 1-20
Porter, M. E., & Kramer, M. R. (2006). Strategy & Society. Harvard Business Review, 84.
Qin, Q., Wang, Q.-G., Li, J., & Ge, S. S. (2013). Linear and nonlinear trading models with gradient boosted random forests and application to Singapore stock market.
Reber, B., Gold, A., & Gold, S. (2021). ESG Disclosure and Idiosyncratic Risk in Ini-tial Public Offerings. Journal of Business Ethics, 1-20.
Rokach, L., & Maimon, O. Z. (2007). Data mining with decision trees: theory and ap-plications (Vol. 69). World scientific.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of in-trinsic motivation, social development, and well-being. American psycholo-gist, 55(1), 68.
Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under condi-tions of risk. The journal of finance, 19(3), 425-442.
Shishkina, A., Dubykovskyy, V., Kosowski, R., & Ramakrishnan, R. (2020). MA-CHINE LEARNING AND RISK-MANAGED INVESTING.
Sunder, S. (2010). Riding the accounting train: from crisis to crisis in eighty years. Pa-per presented at the Presentation at the Conference on Financial Reporting, Audit-ing and Governance, Lehigh University, Bethlehem, PA.
Tan, Z., Yan, Z., & Zhu, G. (2019). Stock selection with random forest: An exploitation of excess return in the Chinese stock market. Heliyon, 5(8), e02310.
Tasnia, M., AlHabshi, S. M. S. J., & Rosman, R. (2020). The impact of corporate so-cial responsibility on stock price volatility of the US banks: A moderating role of tax. Journal of Financial Reporting and Accounting.
van Doorn, J., Onrust, M., Verhoef, P. C., & Bügel, M. S. (2017). The impact of cor-porate social responsibility on customer attitudes and retention—the moderating role of brand success indicators. Marketing Letters, 28(4), 607-619.
Wahba, H. (2008). Does the market value corporate environmental responsibility? An empirical examination. Corporate Social Responsibility and Environmental Man-agement, 15(2), 89-99.
World Economic Forum. (2020). Global Risk Report 2020. http://www3.weforum.org/docs/WEF_Global_Risk_Report_2020.pdf
Yellen, J. L. (1984). Efficiency wage models of unemployment. The american economic review, 74(2), 200-205.
Zhou, F., Zhang, Q., Sornette, D., & Jiang, L. (2019). Cascading logistic regression onto gradient boosted decision trees for forecasting and trading stock indices. Ap-plied Soft Computing, 84, 105747.
描述 碩士
國立政治大學
金融學系
108352001
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0108352001
資料類型 thesis
dc.contributor.advisor 楊曉文zh_TW
dc.contributor.advisor Yang, Sharon. S.en_US
dc.contributor.author (Authors) 孫嘉蔚zh_TW
dc.contributor.author (Authors) Sun, Chia-Weien_US
dc.creator (作者) 孫嘉蔚zh_TW
dc.creator (作者) Sun, Chia-Weien_US
dc.date (日期) 2021en_US
dc.date.accessioned 1-Oct-2021 10:03:16 (UTC+8)-
dc.date.available 1-Oct-2021 10:03:16 (UTC+8)-
dc.date.issued (上傳時間) 1-Oct-2021 10:03:16 (UTC+8)-
dc.identifier (Other Identifiers) G0108352001en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/137284-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 金融學系zh_TW
dc.description (描述) 108352001zh_TW
dc.description.abstract (摘要) 本研究採用機器學習模型,以模型找出何種ESG因子對於台灣企業的公司風險與股價崩跌有較高的解釋力,拆分企業社會責任對於風險的影響。本次研究用梯度提升決策樹(Gradient Boost Trees,以下稱GDBT)、XGBoost、隨機森林(Random Forest),並使用Refinitiv資料庫中的綜合分數、環境分數、社會分數與公司治理分數下的14個指標作為ESG變數。我採用公司特有風險與兩項股價崩跌風險指標(NCSKEW、DUVOL)作為風險變數,以2010-2019年間的992筆台灣公司股價資料計算而成,期望能探究機器學習模型在ESG評分與公司個別風險的效果。
而實證結果顯示,使用XGBoost模型與GDBT在對公司風險與股價崩跌風險模型解釋力上比起隨機森林有較佳的表現。進一步透過分析因子重要性後,數據結果顯示ESG分數綜合指標如ESG綜合分數在公司風險的重要度表現較不佳,顯示比起採用社會責任細項指標,投資人若想依照綜合分數作為投資組合風險管理考量,較無效率。社會類別如企業社會責任策略分數、社區分數與人權分數因子中,在公司特有風險與股價崩跌風險當中,皆具有一定程度的影響性,可以作為企業內部風險管理考量上的指標依據。
zh_TW
dc.description.abstract (摘要) This study approaches three machine learning models to find out which ESG perfor-mance factors have significant impact on firm specific risk and stock crash risks ( NCSKEW,DUVOL ). Three models such as Gradient Boost Trees (Gradient Boost Trees, hereinafter referred to as GDBT), XGBoost, and Random Forest are applied into analyzing the effect of 14 ESG Performance from Reuters database on firm specific risks and stock crash risks. The sample of the study is mainly based Taiwanese firms in Refinitiv ESG database, ranging from the period of 2010 to 2019.The empirical results show that XGBoost and GDBT have the better performance than Random Forest in ex-plaining the company risk and stock crash risk. Through factor importance analysis, I found that combined ESG score are less important in the part of firm risks. This shows that rather than taking ESG composite score into account, investors should consider in-dividual dimensions of Environmental, Social, and Governance indicators for further portfolio risk management.
Social categories, such as corporate social responsibility strategy scores, communi-ty scores, and human rights score, have a certain degree of influence in the firm specific risks and stock crash risks. These scores could be indicators for internal risk manage-ment on the scope of portfolio management and firm risk management.
en_US
dc.description.tableofcontents 第一章 緒論 6
第一節 研究背景及動機 6
第二節 研究目的與研究架構 8
第二章 文獻回顧 10
第一節 企業社會責任文獻探討 10
第二節 企業社會責任與企業風險文獻探討 12
第四節 機器學習文獻探討 14
第三章 樣本選擇與研究方法 18
第一節 機器學習模型說明 18
第二節 風險變數說明 23
第三節 ESG因子變數說明 25
第四節 衡量指標說明 28
第四章 實證結果分析 30
第一節 資料來源與統計分析 30
第二節 機器學習模型績效評估結果 35
第三節 機器學習模型因子重要性分析 37
第五章 結論與未來研究方向 43
參考文獻 46
zh_TW
dc.format.extent 1957091 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0108352001en_US
dc.subject (關鍵詞) 機器學習zh_TW
dc.subject (關鍵詞) ESGzh_TW
dc.subject (關鍵詞) 公司風險zh_TW
dc.subject (關鍵詞) 股價崩跌風險zh_TW
dc.subject (關鍵詞) Machine Learningen_US
dc.subject (關鍵詞) ESGen_US
dc.subject (關鍵詞) Firm Risken_US
dc.subject (關鍵詞) Stock Crash Risken_US
dc.title (題名) 運用機器學習模型分析影響公司風險的ESG因子:以台灣市場為例zh_TW
dc.title (題名) Machine Learning application on the ESG factor analysis on Firm Risksen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) 郭怡萱(2018)。ESG績效對公司風險之影響。國立政治大學財務管理所碩士論文,台北市。
國立臺北大學. (2021). 台灣永續投資調查. http://www.aacsb.ntpu.edu.tw/twsvi/uploads/file/cus2_zu55548c6z.pdf
Annisa, A. N., & Hartanti, D. (2021). The Impact of Environmental, Social, and Gov-ernance Performance on Firm Risk in the ASEAN-5 Countries, 2011-2017. In Asia-Pacific Research in Social Sciences and Humanities Universitas Indonesia Conference (APRISH 2019) (pp. 625-634). Atlantis Press.
Antoncic, M., Bekaert, G., Rothenberg, R. V., & Noguer, M. (2020). Sustainable In-vestment-Exploring the Linkage between Alpha, ESG, and SDG`s. ESG, and SDG`s (August 2020).
Ballings, M., Van den Poel, D., Hespeels, N., & Gryp, R. (2015). Evaluating multiple classifiers for stock price direction prediction. Expert systems with applications, 42(20), 7046-7056.
Broadstock, D. C., Chan, K., Cheng, L. T., & Wang, X. (2021). The role of ESG per-formance during times of financial crisis: Evidence from COVID-19 in China. Fi-nance research letters, 38, 101716.
Chen, J., Hong, H., & Stein, J. C. (2001). Forecasting crashes: Trading volume, past returns, and conditional skewness in stock prices. Journal of financial economics, 61(3), 345-381.
Chen, T., & Guestrin, C. (2016, August). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).
Chen, Q., & Liu, X. Y. (2020, May). Quantifying ESG Alpha in Scholar Big Data: An Automated Machine Learning Approach. In ACM International Conference on AI in Finance.
Crane, A., Matten, D., & Moon, J. (2008). Corporations and citizenship: Business, re-sponsibility and society. Cambridge University Press.
De Lucia, C., Pazienza, P., & Bartlett, M. (2020). Does good ESG lead to better finan-cial performances by firms? Machine learning and logistic regression models of public enterprises in Europe. Sustainability, 12(13), 5317.
Edmans, A. (2011). Does the stock market fully value intangibles? Employee satisfac-tion and equity prices. Journal of Financial economics, 101(3), 621-640.
Ferriani, F., & Natoli, F. (2020). ESG risks in times of COVID-19. Applied Econom-ics Letters, 1-5.
Friedman, J. H. (2001). Greedy function approximation: a gradient boosting ma-chine. Annals of statistics, 1189-1232.
Godfrey, P. C. (2005). The relationship between corporate philanthropy and sharehold-er wealth: A risk management perspective. Academy of management review, 30(4), 777-798.
Godfrey, P. C., Merrill, C. B., & Hansen, J. M. (2009). The relationship between cor-porate social responsibility and shareholder value: An empirical test of the risk management hypothesis. Strategic management journal, 30(4), 425-445.
Guo, T. (2020). Esg2risk: A deep learning framework from esg news to stock volatility prediction. Available at SSRN 3593885.
Harvey, C. R., & Siddique, A. (2000). Conditional skewness in asset pricing tests. The Journal of finance, 55(3), 1263-1295.
Ho, T. K. (1995, August). Random decision forests. In Proceedings of 3rd international conference on document analysis and recognition (Vol. 1, pp. 278-282). IEEE.
Huseynov, F., & Klamm, B. K. (2012). Tax avoidance, tax management and corporate social responsibility. Journal of Corporate Finance, 18(4), 804-827.
Ilhan, E., Sautner, Z., & Vilkov, G. (2021). Carbon tail risk. The Review of Financial Studies, 34(3), 1540-1571.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Jo, H., & Na, H. (2012). Does CSR reduce firm risk? Evidence from controversial in-dustry sectors. Journal of Business Ethics, 110(4), 441-456.
Khaidem, L., Saha, S., & Dey, S. R. (2016). Predicting the direction of stock market prices using random forest. arXiv preprint arXiv:1605.00003.
Kim, Y., Li, H., & Li, S. (2014). Corporate social responsibility and stock price crash risk. Journal of Banking & Finance, 43, 1-13.
KPMG. (2020). The KPMG Survey of Sustainability Reporting 2020. https://assets.kpmg/content/dam/kpmg/xx/pdf/2020/11/the-time-has-come.pdf
Liu, H. (2020). Stock Selection Strategy Based on Support Vector Machine and eX-treme Gradient Boosting Methods. 2020 the 4th International Conference on Big Data Research, 36-39.
Margot, V., Geissler, C., de Franco, C., & Monnier, B. (2021). ESG Investments: Fil-tering versus Machine Learning Approaches. Applied Economics and Finance, 8(2), 1-16.
Minor, D., & Morgan, J. (2011). CSR as reputation insurance: Primum non nocere. California Management Review, 53(3), 40-59.
Mitsuzuka, K., Ling, F., & Ohwada, H. (2017, February). Analysis of CSR activities affecting corporate value using machine learning. In Proceedings of the 9th Inter-national Conference on Machine Learning and Computing (pp. 11-14).
Murata, Hamori, 2021. ESG Disclosures and Stock Price Crash Risk. Journal of Risk and Financial Management, vol. 14(2), pages 1-20
Porter, M. E., & Kramer, M. R. (2006). Strategy & Society. Harvard Business Review, 84.
Qin, Q., Wang, Q.-G., Li, J., & Ge, S. S. (2013). Linear and nonlinear trading models with gradient boosted random forests and application to Singapore stock market.
Reber, B., Gold, A., & Gold, S. (2021). ESG Disclosure and Idiosyncratic Risk in Ini-tial Public Offerings. Journal of Business Ethics, 1-20.
Rokach, L., & Maimon, O. Z. (2007). Data mining with decision trees: theory and ap-plications (Vol. 69). World scientific.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of in-trinsic motivation, social development, and well-being. American psycholo-gist, 55(1), 68.
Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under condi-tions of risk. The journal of finance, 19(3), 425-442.
Shishkina, A., Dubykovskyy, V., Kosowski, R., & Ramakrishnan, R. (2020). MA-CHINE LEARNING AND RISK-MANAGED INVESTING.
Sunder, S. (2010). Riding the accounting train: from crisis to crisis in eighty years. Pa-per presented at the Presentation at the Conference on Financial Reporting, Audit-ing and Governance, Lehigh University, Bethlehem, PA.
Tan, Z., Yan, Z., & Zhu, G. (2019). Stock selection with random forest: An exploitation of excess return in the Chinese stock market. Heliyon, 5(8), e02310.
Tasnia, M., AlHabshi, S. M. S. J., & Rosman, R. (2020). The impact of corporate so-cial responsibility on stock price volatility of the US banks: A moderating role of tax. Journal of Financial Reporting and Accounting.
van Doorn, J., Onrust, M., Verhoef, P. C., & Bügel, M. S. (2017). The impact of cor-porate social responsibility on customer attitudes and retention—the moderating role of brand success indicators. Marketing Letters, 28(4), 607-619.
Wahba, H. (2008). Does the market value corporate environmental responsibility? An empirical examination. Corporate Social Responsibility and Environmental Man-agement, 15(2), 89-99.
World Economic Forum. (2020). Global Risk Report 2020. http://www3.weforum.org/docs/WEF_Global_Risk_Report_2020.pdf
Yellen, J. L. (1984). Efficiency wage models of unemployment. The american economic review, 74(2), 200-205.
Zhou, F., Zhang, Q., Sornette, D., & Jiang, L. (2019). Cascading logistic regression onto gradient boosted decision trees for forecasting and trading stock indices. Ap-plied Soft Computing, 84, 105747.
zh_TW
dc.identifier.doi (DOI) 10.6814/NCCU202101589en_US