人們對於永續的意識不斷提高，ESG指標也成為投資的重要考量因素。因此，本研究使用2012年至2021年台股上市公司之股價資訊、技術指標以及ESG評分資料，首先使用離散小波轉換去除股價噪音，透過基因演算法之特徵篩選技術挑選合適的特徵，並結合極限學習機以預測股價，保留高報酬的股票，以切線及等權重的方法建立投資組合，並進行回測。本文分為A測試、B測試及C測試三種測試，A測試在特徵值篩選階段及加入ESG評分，B測試在機器學習階段加入ESG評分，C測試整個流程都沒有加入ESG分評分。實驗結果發現，經過特徵值篩選後，在A測試和C測試回測表現上較好，其中A測試的波動又比C測試低，故推測ESG具有穩定投資組合波動度效果, so it is speculated that ESG can stabilize the volatility of the investment portfolio。另外，ESG評分中，影響台灣企業股票報酬率最多的為社會責任(S)。

The increasing awareness of sustainable development issues makes ESG indicators become an important consideration for investment. Therefore, this study uses the stock price information, technical indicators and ESG scores of listed companies in Taiwan from 2012 to 2021 as variable data. First, use the discrete wavelet transform to remove the noise of the stock price. Then these data are selected through the feature screening technology of Genetic Algorithm (GA) to select appropriate features and combined with Extreme Learning Machine (ELM) to predict stock prices. Then, according to the forecast results, the high-return stocks are reserved, and the investment portfolio is established with the method of Tangency Portfolio and Equal Weight Portfolio, and the backtest is conducted. This study is divided into three tests: Test A, Test B and Test C. Test A adds ESG scores in the feature selection stage, Test B adds ESG scores in the machine learning stage, Test C does not add ESG scores in the entire process. The experimental results show that after feature selection, the backtest performances of Test A and Test C are better, and the volatility of Test A is lower than that of Test C, so it is speculated that ESG can stabilize the volatility of the investment portfolio. In addition, among the ESG ratings, Social (S) has the greatest impact on the stock returns of Taiwanese companies.

錢慧娟. (2022). 訊號分解對於集成學習預測股價準確率之影響—以台灣加權股價指數為例. 國立政治大學碩士論文.
Alpaydin, E. (2014). Introduction to Machine Learning. MIT Press.
Babatunde, O. H., Armstrong , L., Leng, J., & Diepeveen, D. (2014). A Genetic Algorithm-Based Feature Selection. International Journal of Electronics Communication and Computer Engineering, 5(4), 899–905.
Bassen, A., & Kovacs, A. M. (2008). Environmental, Social and Governance Key Performance Indicators from a Capital Market Perspective.Zeitschrift Für Wirtschaft-Und Unternehmensethik, 9(2), 182–193.
Chandar, S. K., Sumathi, M., & Sivanandam, S. N. (2016). Prediction of Stock Market Price Using Hybrid of Wavelet Transform and Artificial Neural Network. Indian Journal of Science and Technology, 9(8).
Cibas, T., Soulie, F. F., Gallinari, P., & Raudys, S. (1996). Variable Selection with Neural Networks. Neurocomputing, 12, 223–248.
Frohlich, H., Chapelle, O., & Scholkopf, B. (2003). Feature Selection for Support Vector Machines by Means of Genetic Algorithm. IEEE International Conference on Tools with Artificial Intelligence, 142–148.
Holland, J. H. (1975). Adaptation in Natural and Artificial Systems : An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. The MIT Press.
Huang, G., zhu, Q., & Siew, C. (2006). Extreme Learning Machine: Theory and Applications. Neurocomputing, 70(1), 489–501.
Huang, Y., Capretz, L. F., & Ho, D. (2021). Machine Learning for Stock Prediction Based on Fundamental Analysis. IEEE Symposium Series on Computational Intelligence.
Jarrah, M., & Salim, N. (2019). A Recurrent Neural Network and a Discrete Wavelet Transform to Predict the Saudi Stock Price Trends. International Journal of Advanced Computer Science and Applications, 10(4), 164–171.
Kohli, P. P. singh, Zargar, S., Arora, S., & Gupta, P. (2019). Stock Prediction Using Machine Learning Algorithms. Applications of Artificial Intelligence Techniques in Engineering, 1, 405–414.
Li, Y., Wu, J., & Bu, H. (2016). When Quantitative Trading Meets Machine Learning: A Pilot Survey. School of Economics and Management. Beihang University.
Lucia, C. D., Pazienza, P., & Bartlett, M. (2020). Does Good ESG Lead to Better Financial Performances by Firms? Machine Learning and Logistic Regression Models of Public Enterprises in Europe.
Mahesh, B. (2018). Machine Learning Algorithms - A Review. International Journal of Science and Research, 381–386.
Malathi, V., Marimuthu, N. S., Baskar, S., & Ramar, K. (2011). Application of Extreme Learning Machine for Series Compensated Transmission Line Protection. Engineering Applications of Artificial Intelligence, 24(5), 880–887.
Markowitz, H. (1956). Portfolio Selection. The Journal of Finance, 7(1), 77–91.
Miche, Y., Sorjamaa, A., Bas, P., Simula, O., Jutten, C., & Lendasse, A. (2011). OP-ELM: Optimally Pruned Extreme Learning Machine. IEEE Transactions on Neural Networks, 21(1).
Oreski, S., Oreski, D., & Oreski, G. (2012). Hybrid System with Genetic Algorithm and Artificial Neural Networks and Its Application to Retail Credit Risk Assessment. Expert Systems with Applications, 39(16).
Ortega, L., & Khashanah, K. (2014). A Neuro-Wavelet Model for the Short-Term Forecasting of High-Frequency Time Series of Stock Returns. Journal of Forecasting, 33(2).
Raymer, M. L., Punch, W. E., Goodman, E. D., Kuhn, L. A., & Jain, A. K. (2000). Application of Extreme Learning Machine for Series Compensated Transmission Line Protection. IEEE Transactions on Evolutionary Computation, 4(2), 164–171.
Samuel, A. L. (1959). Some Studies in Machine Learning Using the Game of Checkers. IBM Journal, 3(3), 210–229.
Saygili, E., Arslan, S., & Birkan, A. O. (2021). ESG Practices and Corporate Financial Performance: Evidence from Borsa Istanbul. Borsa Istanbul Review, 22(3), 525–533.
Siedlecki, W., & sklansky, J. (1989). A Note on Genetic Algorithms for Large-Scale Feature Selection. Pattern Recognition Letters, 10, 335–347.
Sun, Z. L., Choi, T. M., Au, K. F., & Yu, Y. (2008). Sales Forecasting Using Extreme Learning Machine with Applications in Fashion Retailing. Decision Support Systems, 46(1), 411–419.
Whelan, T., Atz, U., clark, C., & CFA. (2021). Adaptation in Natural and Artificial Systems : An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. NYU STERN.