學術產出-Theses
Article View/Open
Publication Export
-
題名 加權範數懲罰函數之實證研究:以新冠肺炎前後期間中國股市為例
An Empirical Study of Performances of the Weighted Norm Penalized MVP: The Case of Chinese Stock Market During the COVID-19 Period作者 趙思煒
ZHAO, Si-Wei貢獻者 顏佑銘
Yen, Yu-Min
趙思煒
ZHAO, Si-Wei關鍵詞 上證50
加權範數懲罰函數
最小變異數投資組合
新冠肺炎
SSE50
Weighted-norm minimum variance portfolio
Minimum variance portfolio
Covid-19日期 2022 上傳時間 1-Jul-2022 15:58:18 (UTC+8) 摘要 2020年初,新冠疫情在中國爆發,疫情升溫也嚴重影響了經濟發展,股市與經濟息息相關,因此也引起了中國股市震蕩。隨著投資組合理論的發展,在注重報酬的同時,投資人也越來越重視對風險的管控。在實證財務中,最小變異數投資組合(MVP)可說是將風險控管納入投資組合建構之始祖。但最小變異數容易出現極端權重的問題,通過施加懲罰範數可以解決這一問題,並且提高投資組合的稀疏性,而發展出加權範數最小變異數投資組合(WNMVP)。本研究以近十年的 A股市場為研究區間,並以上證 50 作為研究樣本,來對比疫情期間和近十年不同投資組合在 A股市場實證上的表現。本研究對比了加權範數最小變異數投資組合(WNMVP)、做空限制下的最小變異數投資組合(NSMVP)、全局最小變異數投資組合(GMVP)和 1/N均等投資組合,並對WNMVP 分別加上高報酬和低報酬限制,以及使用三種替代性懲罰函數來對比實證結果。本次實證結果證明,在全樣本期間和疫情期間,1/N投資組合雖然在報酬方面表示最佳,但 WNMVP 在報酬方面僅次於 1/N投資組合,且 WNMVP 的風險更低,因此在全樣本期間和疫情期間,WNMVP 都擁有最高的夏普比率。且在疫情期間,WNMVP 在各項衡量指標的表現都更為突出。此外,在全樣本期間和對於施加要求報酬限制式的 WNMVP 實證結果與過往文獻一致,施加要求報酬後的 WNMVP 表現不如無限制下的 WNMVP,而在高要求報酬限制下的WNMVP 實證表現則比在低要求報酬限制下的 WNMVP 更好。
At the beginning of 2020, China witnessed an unprecedented pandemic, COVID-19, scaling up from Wuhan, and it deteriorates the development of domestic economy and caused fluctuation in Chinese stock market. With the development of portfoliotheory, investors are also paying more and more attention to risk managementwhile return on investment remain significant. The application of minimum variance portfolios is becoming more and more extensive in real-world cases. Since the minimum variance portfolio is prone to extreme portfolio weights, imposing a penalty norm can solve this problem and improve the sparsity of the portfolio. As a result, weighted norm minimum variance portfolio (WNMVP) was developed.This study targets the decade of China A-share market and uses the Shanghai 50 Stock Index Futures (SSE50) as sample to compare the empirical performance of different investment portfolios in the China A-share market during the time ofCOVID-19 and in the past ten years. This study compares the weighted norm minimum variance portfolio (WNMVP), the no-shortsale minimum variance portfolio (NSMVP), global minimum variance portfolio (GMVP), and the 1/N equal portfolio,respectively. In addition, this study impose target return constraint into WNMVP, and use three alternative penalty functions to compare empirical results.This empirical result proves that during the full sample period and the period of pandemic, the 1/N portfolio is the best portfolio in terms of return, followed by WNMVP with a lower risk in performance. In conclusion, WNMVP had the highestSharpe ratio in both pandemic period and the last ten years. And WNMVP is recognized to outperform on several performance measures in the time of pandemic, comparing with the last decade. As the previous literature proved, imposing target return constraint into WNMVP deteriorates empirical performance. WNMVP with high-return constraint performs better than the WNMVP with low-return constraint.參考文獻 1. Markowitz, H. (1952). Portfolio Selection.The Journal of Finance, 7, 77–91.2. Merton, R. C. (1980). On Estimating The Expected Return On The Market: AnExploratory Investigation, Journal of Financial Economics,8(4), 323-3613. DeMiguel, V., Garlappi, L., Nogales, F. J., and Uppal, R. (2009a) A generalized approach to portfolio optimization: improving performance by constraining portfolio norms, Management Science55, 798–812.4. Haugen, R. A., and Baker, N. L. (1991). The Efficient Market Inefficiency of Capitalization-Weighted Stock Portfolios. Journal of Portfolio Management, vol. 17,no. 3, spring: 35-40.5. Jagannathan, R. and Ma, T., (2003). Risk reduction in large portfolios: why imposing the wrong constraints helps, Journal of Finance 58, 1651–1684.6. Brodie, J., Daubechies, I., De Mol, C., Giannone, D., and Loris, I. (2009). Sparse and stable Markowitz portfolios. Proceedings of the National Academy of Sciences of the United States of America, 106(30), 12267–12272.7. Ledoit, O. and Wolf, M. (2003) Improved estimation of the covariance matrix of stock returns with an application to portfolio selection, Journal of Empirical Finance 10, 603–621.8. Chen, A. Pong, E and Wang, Y. (2018). Accessing the China A-Shares Market via Minimum-Variance Investing. The Journal of Portfolio Management Fall 2018, 45 (1) 106-117;9. Yen, Y. M. (2015). Sparse Weighted-Norm Minimum Variance Portfolios. Review of Finance, 20, 1259-1287.10. Jorion, P.(1986). Bayes-Stein Estimation for Portfolio Analysis. Journal of Financial and Quantitative Analysis,21(3), 279 – 29211.Clarke, R.,Desilva, H. and Thorley, S.(2006) Long–Short Extensions: How Much Is Enough?. Portfolio Management, Volume 64, 2008 - Issue 1,16-30 3612. Haugen, R. A., and Baker, N. L.(2012). Low Risk Stocks Outperform within All Observable Markets of the World.13. DeMiguel, V., Garlappi, L., and Uppal, R. (2009b) Optimal versus naive diversification: how inefficient is the 1/N portfolio strategy?, Review of Financial Studies 22, 1915–1953.14. Owen, A. B. (2007) A robust hybrid of lasso and ridge regression, ContemporaryMathematics 443, 59–72.15. DeMiguel, V., Nogales, F. J., and Uppal, R. (2014) Stock return serial dependence and OOS portfolio performance, Review of Financial Studies 27, 1031–107316. Gabaix, X. (2015) A sparsity-based model of bounded rationality, Quarterly Journal of Economics 129, 1661–1710.17. Friedman, J., Hastie, T., Hofling, H., and Tibshirani, R. (2007) Pathwise coordinate optimization, Annals of Applied Statistics 1, 302–332.18. 徐宏,蒲紅霞(2021)。新冠疫情對中國股票市場的影響——基於事件研究法的研究。《金融論壇》,26(7),70-80 描述 碩士
國立政治大學
國際經營與貿易學系
108351050資料來源 http://thesis.lib.nccu.edu.tw/record/#G0108351050 資料類型 thesis dc.contributor.advisor 顏佑銘 zh_TW dc.contributor.advisor Yen, Yu-Min en_US dc.contributor.author (Authors) 趙思煒 zh_TW dc.contributor.author (Authors) ZHAO, Si-Wei en_US dc.creator (作者) 趙思煒 zh_TW dc.creator (作者) ZHAO, Si-Wei en_US dc.date (日期) 2022 en_US dc.date.accessioned 1-Jul-2022 15:58:18 (UTC+8) - dc.date.available 1-Jul-2022 15:58:18 (UTC+8) - dc.date.issued (上傳時間) 1-Jul-2022 15:58:18 (UTC+8) - dc.identifier (Other Identifiers) G0108351050 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/140545 - 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) 108351050 zh_TW dc.description.abstract (摘要) 2020年初,新冠疫情在中國爆發,疫情升溫也嚴重影響了經濟發展,股市與經濟息息相關,因此也引起了中國股市震蕩。隨著投資組合理論的發展,在注重報酬的同時,投資人也越來越重視對風險的管控。在實證財務中,最小變異數投資組合(MVP)可說是將風險控管納入投資組合建構之始祖。但最小變異數容易出現極端權重的問題,通過施加懲罰範數可以解決這一問題,並且提高投資組合的稀疏性,而發展出加權範數最小變異數投資組合(WNMVP)。本研究以近十年的 A股市場為研究區間,並以上證 50 作為研究樣本,來對比疫情期間和近十年不同投資組合在 A股市場實證上的表現。本研究對比了加權範數最小變異數投資組合(WNMVP)、做空限制下的最小變異數投資組合(NSMVP)、全局最小變異數投資組合(GMVP)和 1/N均等投資組合,並對WNMVP 分別加上高報酬和低報酬限制,以及使用三種替代性懲罰函數來對比實證結果。本次實證結果證明,在全樣本期間和疫情期間,1/N投資組合雖然在報酬方面表示最佳,但 WNMVP 在報酬方面僅次於 1/N投資組合,且 WNMVP 的風險更低,因此在全樣本期間和疫情期間,WNMVP 都擁有最高的夏普比率。且在疫情期間,WNMVP 在各項衡量指標的表現都更為突出。此外,在全樣本期間和對於施加要求報酬限制式的 WNMVP 實證結果與過往文獻一致,施加要求報酬後的 WNMVP 表現不如無限制下的 WNMVP,而在高要求報酬限制下的WNMVP 實證表現則比在低要求報酬限制下的 WNMVP 更好。 zh_TW dc.description.abstract (摘要) At the beginning of 2020, China witnessed an unprecedented pandemic, COVID-19, scaling up from Wuhan, and it deteriorates the development of domestic economy and caused fluctuation in Chinese stock market. With the development of portfoliotheory, investors are also paying more and more attention to risk managementwhile return on investment remain significant. The application of minimum variance portfolios is becoming more and more extensive in real-world cases. Since the minimum variance portfolio is prone to extreme portfolio weights, imposing a penalty norm can solve this problem and improve the sparsity of the portfolio. As a result, weighted norm minimum variance portfolio (WNMVP) was developed.This study targets the decade of China A-share market and uses the Shanghai 50 Stock Index Futures (SSE50) as sample to compare the empirical performance of different investment portfolios in the China A-share market during the time ofCOVID-19 and in the past ten years. This study compares the weighted norm minimum variance portfolio (WNMVP), the no-shortsale minimum variance portfolio (NSMVP), global minimum variance portfolio (GMVP), and the 1/N equal portfolio,respectively. In addition, this study impose target return constraint into WNMVP, and use three alternative penalty functions to compare empirical results.This empirical result proves that during the full sample period and the period of pandemic, the 1/N portfolio is the best portfolio in terms of return, followed by WNMVP with a lower risk in performance. In conclusion, WNMVP had the highestSharpe ratio in both pandemic period and the last ten years. And WNMVP is recognized to outperform on several performance measures in the time of pandemic, comparing with the last decade. As the previous literature proved, imposing target return constraint into WNMVP deteriorates empirical performance. WNMVP with high-return constraint performs better than the WNMVP with low-return constraint. en_US dc.description.tableofcontents 第一章 緒論 6第一節 研究動機及目的 6第二節 研究架構 8第二章 文獻探討10第一節 投資組合理論起源 10第二節 不同投資組合理論對比 11第三節 懲罰範數投資策略 12第三章 研究方法14第一節 建構並求解 WNMVP 最適化 14第二節 用於對比的其他投資組合標桿 16第三節 績效衡量指標說明 17第四節 其他替代性懲罰函數 20第四章 實證分析22第一節 樣本資料敘述 22第二節 實證結果分析 26第三節 替代性懲罰函數實證分析 31第五章 研究結論及建議33參考文獻 35 zh_TW dc.format.extent 1712140 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0108351050 en_US dc.subject (關鍵詞) 上證50 zh_TW dc.subject (關鍵詞) 加權範數懲罰函數 zh_TW dc.subject (關鍵詞) 最小變異數投資組合 zh_TW dc.subject (關鍵詞) 新冠肺炎 zh_TW dc.subject (關鍵詞) SSE50 en_US dc.subject (關鍵詞) Weighted-norm minimum variance portfolio en_US dc.subject (關鍵詞) Minimum variance portfolio en_US dc.subject (關鍵詞) Covid-19 en_US dc.title (題名) 加權範數懲罰函數之實證研究:以新冠肺炎前後期間中國股市為例 zh_TW dc.title (題名) An Empirical Study of Performances of the Weighted Norm Penalized MVP: The Case of Chinese Stock Market During the COVID-19 Period en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) 1. Markowitz, H. (1952). Portfolio Selection.The Journal of Finance, 7, 77–91.2. Merton, R. C. (1980). On Estimating The Expected Return On The Market: AnExploratory Investigation, Journal of Financial Economics,8(4), 323-3613. DeMiguel, V., Garlappi, L., Nogales, F. J., and Uppal, R. (2009a) A generalized approach to portfolio optimization: improving performance by constraining portfolio norms, Management Science55, 798–812.4. Haugen, R. A., and Baker, N. L. (1991). The Efficient Market Inefficiency of Capitalization-Weighted Stock Portfolios. Journal of Portfolio Management, vol. 17,no. 3, spring: 35-40.5. Jagannathan, R. and Ma, T., (2003). Risk reduction in large portfolios: why imposing the wrong constraints helps, Journal of Finance 58, 1651–1684.6. Brodie, J., Daubechies, I., De Mol, C., Giannone, D., and Loris, I. (2009). Sparse and stable Markowitz portfolios. Proceedings of the National Academy of Sciences of the United States of America, 106(30), 12267–12272.7. Ledoit, O. and Wolf, M. (2003) Improved estimation of the covariance matrix of stock returns with an application to portfolio selection, Journal of Empirical Finance 10, 603–621.8. Chen, A. Pong, E and Wang, Y. (2018). Accessing the China A-Shares Market via Minimum-Variance Investing. The Journal of Portfolio Management Fall 2018, 45 (1) 106-117;9. Yen, Y. M. (2015). Sparse Weighted-Norm Minimum Variance Portfolios. Review of Finance, 20, 1259-1287.10. Jorion, P.(1986). Bayes-Stein Estimation for Portfolio Analysis. Journal of Financial and Quantitative Analysis,21(3), 279 – 29211.Clarke, R.,Desilva, H. and Thorley, S.(2006) Long–Short Extensions: How Much Is Enough?. Portfolio Management, Volume 64, 2008 - Issue 1,16-30 3612. Haugen, R. A., and Baker, N. L.(2012). Low Risk Stocks Outperform within All Observable Markets of the World.13. DeMiguel, V., Garlappi, L., and Uppal, R. (2009b) Optimal versus naive diversification: how inefficient is the 1/N portfolio strategy?, Review of Financial Studies 22, 1915–1953.14. Owen, A. B. (2007) A robust hybrid of lasso and ridge regression, ContemporaryMathematics 443, 59–72.15. DeMiguel, V., Nogales, F. J., and Uppal, R. (2014) Stock return serial dependence and OOS portfolio performance, Review of Financial Studies 27, 1031–107316. Gabaix, X. (2015) A sparsity-based model of bounded rationality, Quarterly Journal of Economics 129, 1661–1710.17. Friedman, J., Hastie, T., Hofling, H., and Tibshirani, R. (2007) Pathwise coordinate optimization, Annals of Applied Statistics 1, 302–332.18. 徐宏,蒲紅霞(2021)。新冠疫情對中國股票市場的影響——基於事件研究法的研究。《金融論壇》,26(7),70-80 zh_TW dc.identifier.doi (DOI) 10.6814/NCCU202200548 en_US