Publications-Theses
Article View/Open
Publication Export
-
Google ScholarTM
NCCU Library
Citation Infomation
Related Publications in TAIR
題名 以機器學習模型建構多空投資組合策略
Constructing Long-Short Investment Portfolio Strategies Using Machine Learning Models作者 黃紀豪
Huang, Ji-Hao貢獻者 鍾令德
Chung, Ling-Tak
黃紀豪
Huang, Ji-Hao關鍵詞 機器學習
股價報酬預測
投資組合選擇
Machine Learning
Return Predictability
Portfolio Choice日期 2024 上傳時間 5-Aug-2024 11:57:07 (UTC+8) 摘要 本研究比較 18 個機器學習模型預測台灣股市上市公司報酬的能力，並回測以 10 種不同多空比例及加權比重建構之 180 種機器學習投資策略。結果顯示 11 種機器學習模型能有效預測個股超額報酬，神經網路模型、決策樹模型預測表現較佳，其中 XGBoost 模型建構之多空投資組合策略績效最為優異，能在樣本外期間獲得 3.58% 之月均報酬，並達到 3.56 之年化夏普比率，顯示機器學習模型確實能捕捉股票特徵與下期報酬的非線性關係，產生有價值之交易訊號，進而為投資人帶來顯著的報酬。另外，本研究發現 10-1 分位多空策略相較全市場多空策略能有效提升夏普比率，而 130/30 策略雖然能創造比淨零投資組合策略更高的報酬，卻因波動性更高而無法有效提升夏普比率。
This study evaluates 18 machine learning models in predicting stock returns of listed companies in Taiwan. Through 10 combinations of long-short ratios and weighting schemes, I backtest 180 investment strategies based on machine learning predictions. The results show that 11 machine learning models can effectively predict individual stock excess returns. Neural network models and decision tree models exhibit better predictive performance, with the XGBoost model constructing the best performing long-short investment portfolio strategy. This strategy achieves an average monthly return of 3.58\% and an annualized Sharpe ratio of 3.56 during the out-of-sample period. Machine learning models can capture non-linear relationships between stock characteristics and future returns, generating valuable trading signals that bring significant Alphas for investors. Furthermore, this study finds that the 10-1 long-short strategy effectively improves the Sharpe ratio compared to full market long-short strategies. Although the 130/30 strategy can generate higher returns than net-zero investment strategies, it fails to effectively improve the Sharpe ratio due to its higher volatility.參考文獻 Alpaydin, Ethem, 2010, Introduction to Machine Learning (MIT Press). Amihud, Yakov, and Haim Mendelson, 1986, Liquidity and Stock Returns, Financial Analysts Journal 42, 43–48. Ang, Andrew, Robert J. Hodrick, Yuhang Xing, and Xiaoyan Zhang, 2006, The Cross- Section of Volatility and Expected Returns, The Journal of Finance 61, 259–299. Bali, Turan G., and Nusret Cakici, 2008, Idiosyncratic Volatility and the Cross Section of Expected Returns, Journal of Financial and Quantitative Analysis 43, 29–58. Bao, Yanlin, 2023, Replication of Gu, Kelly and Xiu (2020, RFS), Re- trieved May 25, 2024, from https://colab.research.google.com/drive/ 1fcWNL5CgD21kuFDRLvYpEV8l9f5m2n9m#scrollTo=463f2485. Blitz, David, Hoogteijling Tobias, Harald Lohre, and Messow Philip, 2023, How Can Machine Learning Advance Quantitative Asset Management?, The Journal of Port- folio Management 50, 31–63. Buchanan, Lauren J., 2011, The Success of Long-Short Equity Strategies versus Tradi- tional Equity Strategies Market Returns. Bui, Dien Giau, De-Rong Kong, Chih-Yung Lin, and Tse-Chun Lin, 2023, Momentum in machine learning: Evidence from the Taiwan stock market, Pacific-Basin Finance Journal 82, 102178. Chen, Andrew Y., and Tom Zimmermann, 2021, Open Source Cross-Sectional Asset Pricing, Critical Financial Review 11, 207–264. Chen, Tianqi, and Carlos Guestrin, 2016, XGBoost: A scalable tree boosting system, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794, ACM. Dopfel, Frederick E., and Sunder R. Ramkumar, 2005, The Efficiency Gains of Long- Short Credit Strategies, The Journal of Fixed Income 15, 5–15. Drobetz, Wolfgang, Fabian Hollstein, Tizian Otto, and Marcel Prokopczuk, 2024, Esti- mating Stock Market Betas via Machine Learning, Journal of Financial and Quanti- tative Analysis 1–56. Fama, Eugene F., and Kenneth R. French, 2015, A Five-Factor Asset Pricing Model, Journal of Financial Economics 116, 1–22. Freyberger, Joachim, Andreas Neuhierl, and Michael Weber, 2020, Dissecting Charac- teristics Nonparametrically, The Review of Financial Studies 33, 2326–2377. Grinold, Richard C., and Ronald N. Kahn, 2000, The Efficiency Gains of Long–Short Investing, Financial Analysts Journal 56, 40–53. Gu, Shihao, Bryan Kelly, and Dacheng Xiu, 2020, Empirical Asset Pricing via Machine Learning, The Review of Financial Studies 33, 2223–2273. Hameed, Allaudeen, and Yuanto Kusnadi, 2002, Momentum Strategies: Evidence from Pacific Basin Stock Markets, The Journal of Financial Research 25, 383–397. Heaton, J.B., Nick Polson, and Jan Witte, 2016, Deep Learning for Finance: Deep Port- folios, Applied Stochastic Models in Business and Industry 33, 3–12. Htun, H.H., M. Biehl, and N. Petkov, 2023, Survey of feature selection and extraction techniques for stock market prediction, Financial Innovation 9. Huber, Peter J, 1964, Robust estimation of a location parameter, The Annals of Mathe- matical Statistics 35, 73–101. Jacobs, Bruce I., and Kenneth N. Levy, 1993, Long/Short Equity Investing, The Journal of Portfolio Management 20, 52–63. Jegadeesh, Narasimhan, and Sheridan Titman, 1993, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, The Journal of Finance 48, 65–91. Kelly, Bryan T., Seth Pruitt, and Yinan Su, 2019, Characteristics are covariances: A unified model of risk and return, Journal of Financial Economics 134, 501–524. Kingma, Diederik P., and Jimmy Ba, 2015, Adam: A Method for Stochastic Optimiza- tion, in Proceedings of the 3rd International Conference for Learning Representa- tions, San Diego. Lee, Yongjae, John R. J. Thompson, Jang Ho Kim, Woo Chang Kim, and Francesco A. Fabozzi, 2023, An Overview of Machine Learning for Asset Management, The Jour- nal of Portfolio Management 49, 31–63. Lewellen, Jonathan, 2015, The Cross-section of Expected Stock Returns, Critical Fi- nance Review 20, 1–44. Miffre, Joëlle, and Adrian Fernandez-Perez, 2015, The Case for Long-Short Commodity Investing, The Journal of Alternative Investments 18, 92–104. Moritz, Benjamin, and Tom Zimmermann, 2016, Tree-Based Conditional Portfolio Sorts: The Relation between Past and Future Stock Returns, Working Paper, Lud- wig Maximilian University of Munich. Pástor, Ľuboš, and Robert F. Stambaugh, 2003, Liquidity Risk and Expected Stock Re- turns, Journal of Political Economy 111, 642–685. Rapach, David, Jack Strauss, and Guofu Zhou, 2012, How Can Machine Learning Ad- vance Quantitative Asset Management?, Journal of Finance 68, 1633–1662. Sadhwani, Apaar, Kay Giesecke, and Justin Sirignano, 2021, Deep Learning for Mort- gage Risk, Journal of Financial Econometrics 19, 313–368. Samuel, A. L., 1959, Some Studies in Machine Learning Using the Game of Checkers, IBM Journal of Research and Development 3, 210–229. Spearman, Charles, 1904, The proof and measurement of association between two things, American Journal of Psychology 15, 72–101. Tol, Ramon, and Christiaan Wanningen, 2009, On the Performance of Extended Alpha (130/30) versus Long-Only, The Journal of Portfolio Management 35, 51–60. Tol, Ramon, and Christiaan Wanningen, 2011, 130/30: By How Much Will the Infor- mation Ratio, The Journal of Portfolio Management 37, 62–69. Tsai, Pei-Fen, Cheng-Han Gao, and Shyan-Ming Yuan, 2023, Stock Selection Using Machine Learning Based on Financial Ratios, Mathematics 11. Turing, Alan M., 1950, Computing Machinery and Intelligence, Mind 59, 433–60. Waid, Robert J, 2009, Long-Only: The Natural Benchmark Choice for 130/30, The Journal of Portfolio Management 35, 48–50. Wang, Yun-Chin, Jean Yu, and Shiow-Ying Wen, 2014, Does Fundamental and Tech- nical Analysis Reduce Investment Risk for Growth Stock? An Analysis of Taiwan Stock Market, International Business Research 7, 24–34. White, 1988, Economic Prediction Using Neural Networks: The Case of IBM Daily Stock Returns, in IEEE 1988 International Conference on Neural Networks, 451– 458 vol.2. 蔣佳穎, 2022, 以財務指標預測台股橫斷面期望報酬 ，未出版之博 (碩) 士論文， 國立政治大學，國際經營與貿易學系，台北市. 描述 碩士
國立政治大學
國際經營與貿易學系
111351024資料來源 http://thesis.lib.nccu.edu.tw/record/#G0111351024 資料類型 thesis dc.contributor.advisor 鍾令德 zh_TW dc.contributor.advisor Chung, Ling-Tak en_US dc.contributor.author (Authors) 黃紀豪 zh_TW dc.contributor.author (Authors) Huang, Ji-Hao en_US dc.creator (作者) 黃紀豪 zh_TW dc.creator (作者) Huang, Ji-Hao en_US dc.date (日期) 2024 en_US dc.date.accessioned 5-Aug-2024 11:57:07 (UTC+8) - dc.date.available 5-Aug-2024 11:57:07 (UTC+8) - dc.date.issued (上傳時間) 5-Aug-2024 11:57:07 (UTC+8) - dc.identifier (Other Identifiers) G0111351024 en_US dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/152398 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 國際經營與貿易學系 zh_TW dc.description (描述) 111351024 zh_TW dc.description.abstract (摘要) 本研究比較 18 個機器學習模型預測台灣股市上市公司報酬的能力，並回測以 10 種不同多空比例及加權比重建構之 180 種機器學習投資策略。結果顯示 11 種機器學習模型能有效預測個股超額報酬，神經網路模型、決策樹模型預測表現較佳，其中 XGBoost 模型建構之多空投資組合策略績效最為優異，能在樣本外期間獲得 3.58% 之月均報酬，並達到 3.56 之年化夏普比率，顯示機器學習模型確實能捕捉股票特徵與下期報酬的非線性關係，產生有價值之交易訊號，進而為投資人帶來顯著的報酬。另外，本研究發現 10-1 分位多空策略相較全市場多空策略能有效提升夏普比率，而 130/30 策略雖然能創造比淨零投資組合策略更高的報酬，卻因波動性更高而無法有效提升夏普比率。 zh_TW dc.description.abstract (摘要) This study evaluates 18 machine learning models in predicting stock returns of listed companies in Taiwan. Through 10 combinations of long-short ratios and weighting schemes, I backtest 180 investment strategies based on machine learning predictions. The results show that 11 machine learning models can effectively predict individual stock excess returns. Neural network models and decision tree models exhibit better predictive performance, with the XGBoost model constructing the best performing long-short investment portfolio strategy. This strategy achieves an average monthly return of 3.58\% and an annualized Sharpe ratio of 3.56 during the out-of-sample period. Machine learning models can capture non-linear relationships between stock characteristics and future returns, generating valuable trading signals that bring significant Alphas for investors. Furthermore, this study finds that the 10-1 long-short strategy effectively improves the Sharpe ratio compared to full market long-short strategies. Although the 130/30 strategy can generate higher returns than net-zero investment strategies, it fails to effectively improve the Sharpe ratio due to its higher volatility. en_US dc.description.tableofcontents 第一章 緒論 1 第一節 研究背景與動機 1 第二節 研究目的 2 第三節 研究架構 2 第二章 文獻回顧 4 第一節 機器學習於資產定價之應用 4 第二節 多空投資組合策略 6 第三節 特徵變數選擇 7 第四節 文獻回顧總結 9 第三章 研究資料與方法 10 第一節 研究對象 10 第二節 模型設定 11 第三節 特徵變數 24 第四節 模型績效評估 26 第四章 研究結果與分析 28 第一節 模型預測評估結果 28 第二節 投資組合策略回測 38 第五章 結論與建議 62 第一節 結論 62 第二節 限制與建議 62 參考文獻 64 附錄 68 zh_TW dc.format.extent 3056457 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0111351024 en_US dc.subject (關鍵詞) 機器學習 zh_TW dc.subject (關鍵詞) 股價報酬預測 zh_TW dc.subject (關鍵詞) 投資組合選擇 zh_TW dc.subject (關鍵詞) Machine Learning en_US dc.subject (關鍵詞) Return Predictability en_US dc.subject (關鍵詞) Portfolio Choice en_US dc.title (題名) 以機器學習模型建構多空投資組合策略 zh_TW dc.title (題名) Constructing Long-Short Investment Portfolio Strategies Using Machine Learning Models en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) Alpaydin, Ethem, 2010, Introduction to Machine Learning (MIT Press). Amihud, Yakov, and Haim Mendelson, 1986, Liquidity and Stock Returns, Financial Analysts Journal 42, 43–48. Ang, Andrew, Robert J. Hodrick, Yuhang Xing, and Xiaoyan Zhang, 2006, The Cross- Section of Volatility and Expected Returns, The Journal of Finance 61, 259–299. Bali, Turan G., and Nusret Cakici, 2008, Idiosyncratic Volatility and the Cross Section of Expected Returns, Journal of Financial and Quantitative Analysis 43, 29–58. Bao, Yanlin, 2023, Replication of Gu, Kelly and Xiu (2020, RFS), Re- trieved May 25, 2024, from https://colab.research.google.com/drive/ 1fcWNL5CgD21kuFDRLvYpEV8l9f5m2n9m#scrollTo=463f2485. Blitz, David, Hoogteijling Tobias, Harald Lohre, and Messow Philip, 2023, How Can Machine Learning Advance Quantitative Asset Management?, The Journal of Port- folio Management 50, 31–63. Buchanan, Lauren J., 2011, The Success of Long-Short Equity Strategies versus Tradi- tional Equity Strategies Market Returns. Bui, Dien Giau, De-Rong Kong, Chih-Yung Lin, and Tse-Chun Lin, 2023, Momentum in machine learning: Evidence from the Taiwan stock market, Pacific-Basin Finance Journal 82, 102178. Chen, Andrew Y., and Tom Zimmermann, 2021, Open Source Cross-Sectional Asset Pricing, Critical Financial Review 11, 207–264. Chen, Tianqi, and Carlos Guestrin, 2016, XGBoost: A scalable tree boosting system, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794, ACM. Dopfel, Frederick E., and Sunder R. Ramkumar, 2005, The Efficiency Gains of Long- Short Credit Strategies, The Journal of Fixed Income 15, 5–15. Drobetz, Wolfgang, Fabian Hollstein, Tizian Otto, and Marcel Prokopczuk, 2024, Esti- mating Stock Market Betas via Machine Learning, Journal of Financial and Quanti- tative Analysis 1–56. Fama, Eugene F., and Kenneth R. French, 2015, A Five-Factor Asset Pricing Model, Journal of Financial Economics 116, 1–22. Freyberger, Joachim, Andreas Neuhierl, and Michael Weber, 2020, Dissecting Charac- teristics Nonparametrically, The Review of Financial Studies 33, 2326–2377. Grinold, Richard C., and Ronald N. Kahn, 2000, The Efficiency Gains of Long–Short Investing, Financial Analysts Journal 56, 40–53. Gu, Shihao, Bryan Kelly, and Dacheng Xiu, 2020, Empirical Asset Pricing via Machine Learning, The Review of Financial Studies 33, 2223–2273. Hameed, Allaudeen, and Yuanto Kusnadi, 2002, Momentum Strategies: Evidence from Pacific Basin Stock Markets, The Journal of Financial Research 25, 383–397. Heaton, J.B., Nick Polson, and Jan Witte, 2016, Deep Learning for Finance: Deep Port- folios, Applied Stochastic Models in Business and Industry 33, 3–12. Htun, H.H., M. Biehl, and N. Petkov, 2023, Survey of feature selection and extraction techniques for stock market prediction, Financial Innovation 9. Huber, Peter J, 1964, Robust estimation of a location parameter, The Annals of Mathe- matical Statistics 35, 73–101. Jacobs, Bruce I., and Kenneth N. Levy, 1993, Long/Short Equity Investing, The Journal of Portfolio Management 20, 52–63. Jegadeesh, Narasimhan, and Sheridan Titman, 1993, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, The Journal of Finance 48, 65–91. Kelly, Bryan T., Seth Pruitt, and Yinan Su, 2019, Characteristics are covariances: A unified model of risk and return, Journal of Financial Economics 134, 501–524. Kingma, Diederik P., and Jimmy Ba, 2015, Adam: A Method for Stochastic Optimiza- tion, in Proceedings of the 3rd International Conference for Learning Representa- tions, San Diego. Lee, Yongjae, John R. J. Thompson, Jang Ho Kim, Woo Chang Kim, and Francesco A. Fabozzi, 2023, An Overview of Machine Learning for Asset Management, The Jour- nal of Portfolio Management 49, 31–63. Lewellen, Jonathan, 2015, The Cross-section of Expected Stock Returns, Critical Fi- nance Review 20, 1–44. Miffre, Joëlle, and Adrian Fernandez-Perez, 2015, The Case for Long-Short Commodity Investing, The Journal of Alternative Investments 18, 92–104. Moritz, Benjamin, and Tom Zimmermann, 2016, Tree-Based Conditional Portfolio Sorts: The Relation between Past and Future Stock Returns, Working Paper, Lud- wig Maximilian University of Munich. Pástor, Ľuboš, and Robert F. Stambaugh, 2003, Liquidity Risk and Expected Stock Re- turns, Journal of Political Economy 111, 642–685. Rapach, David, Jack Strauss, and Guofu Zhou, 2012, How Can Machine Learning Ad- vance Quantitative Asset Management?, Journal of Finance 68, 1633–1662. Sadhwani, Apaar, Kay Giesecke, and Justin Sirignano, 2021, Deep Learning for Mort- gage Risk, Journal of Financial Econometrics 19, 313–368. Samuel, A. L., 1959, Some Studies in Machine Learning Using the Game of Checkers, IBM Journal of Research and Development 3, 210–229. Spearman, Charles, 1904, The proof and measurement of association between two things, American Journal of Psychology 15, 72–101. Tol, Ramon, and Christiaan Wanningen, 2009, On the Performance of Extended Alpha (130/30) versus Long-Only, The Journal of Portfolio Management 35, 51–60. Tol, Ramon, and Christiaan Wanningen, 2011, 130/30: By How Much Will the Infor- mation Ratio, The Journal of Portfolio Management 37, 62–69. Tsai, Pei-Fen, Cheng-Han Gao, and Shyan-Ming Yuan, 2023, Stock Selection Using Machine Learning Based on Financial Ratios, Mathematics 11. Turing, Alan M., 1950, Computing Machinery and Intelligence, Mind 59, 433–60. Waid, Robert J, 2009, Long-Only: The Natural Benchmark Choice for 130/30, The Journal of Portfolio Management 35, 48–50. Wang, Yun-Chin, Jean Yu, and Shiow-Ying Wen, 2014, Does Fundamental and Tech- nical Analysis Reduce Investment Risk for Growth Stock? An Analysis of Taiwan Stock Market, International Business Research 7, 24–34. White, 1988, Economic Prediction Using Neural Networks: The Case of IBM Daily Stock Returns, in IEEE 1988 International Conference on Neural Networks, 451– 458 vol.2. 蔣佳穎, 2022, 以財務指標預測台股橫斷面期望報酬 ，未出版之博 (碩) 士論文， 國立政治大學，國際經營與貿易學系，台北市. zh_TW
