學術產出-Theses
Article View/Open
Publication Export
-
題名 運用深度強化學習建立虛擬貨幣投資組合
Establish The Portfolio of Crypto Currency by Applying Deep-Reinforcement Learning作者 蘇育正
Su, Yu-Cheng貢獻者 蔡炎龍<br>蕭明福
Thai, YenLung<br>Shaw, MingFu
蘇育正
Su, Yu-Cheng關鍵詞 深度學習
強化學習
深度強化學習
虛擬貨幣
投資組合
Reinforcement Learning
Portfolio
Crypto日期 2022 上傳時間 9-Mar-2023 18:23:05 (UTC+8) 摘要 本研究運用深度強化學習建立虛擬貨幣的投資組合，研究標的主要以 2021年 12 月 31 日市值排名前 50 大的虛擬貨幣。研究期間從 2017 年 1 月 3 日至2021 年 12 月 31 日，並主要以五個因子(Factor):開盤價(Open)、最高價(High)、最低價(Low)、收盤價(Close)、成交量(Volume)為輸入資料(Input)，並在一開始先以(1)市值、(2)平均振幅抓取 30 檔虛擬幣組建投資組合，輸入給深度強化學習模型進行訓練，最終發現相較於其他種因子建立的投資組合，平均振幅打造的投資組合表現更好，也比單一持續持有比特幣來的更合適。 參考文獻 [1] Fan Fang, Carmine Ventre, Michail Basios, Leslie Kanthan, DavidMartinez-Rego, Fan Wu and Lingbo Li. Cryptocurrency trading: a comprehensive survey. Finanical Innovation,8(13),2022.[2] Timothy King and Dimitrios Koutmos. Herding and feedback trading in cryptocurrency markets. Annals of Operations Research,300:79-97,2021.[3] WeiSun, Alisher Tohirovich, Dedahanov, Ho YoungShin and Wei PingLi. Factors affecting institutional investors to add cryptocurrency to asset portfolios. The North American Journal of Economics and Finance,volume 58,2021.[4] Paraskevi Katsiampa, Larisa Yarovaya and DamianZięba. Highfrequency connectedness between Bitcoin and other top-traded crypto assets during the COVID-19 crisis. Journal of International Financial Markets, Institutions and Money,volume 79,2022.[5] Andrés Arévalo, Jaime Niño, G. Hernández and Javier Sandoval. High-Frequency Trading Strategy Based on Deep Neural Networks.International Conference on intelligent Computing, LNAI,volume 9773,2016.[6] Maria Čuljak,BojanTomić and SašaŽiković . Benefits of sectoral cryptocurrency portfolio optimization. Research in International Business and Finance,volume 60,2022.[7] Golnoosh Babaei,Paolo Giudici and EmanuelaRaffinetti. Explainable artificial intelligence for crypto asset allocation. Finance Research Letters,volume 47,Part B,2022.[8] Leonardo Kanashiro Felizardo,Francisco CaioLima Paiva,Catharinede Vita Graves,Elia Yathie Matsumoto,Anna Helena Reali Costa,Emilio DelMoral-Hernandez and Paolo Brandimarte. Outperforming algorithmic trading reinforcement learning systems: A supervised approach to the 30cryptocurrency market. Expert Systems with Applications,volume 202,2022.[10] Hongfeng Xu,Lei Chai,Zhiming Luo and Shaozi Li. Stock movement prediction via gated recurrent unit network based on reinforcement learning with incorporated attention mechanisms. Neurocomputing,volume 467,Pages 214-228,2022.[11] Fengrui Liu,Yang Li,Baitong Li,Jiaxin Li and Huiyang Xie . Bitcoin transaction strategy construction based on deep reinforcement learning. Applied Soft Computing,volume113,Part B,2021.[12] Thibaut Théate and Damien Ernst . An application of deep reinforcement learning to algorithmic trading. Expert Systems with Applications,volume 173,2021.[13] Liguo Weng,Xudong Sun,Min Xia,Jia Liu and Yiqing Xu. Portfolio Trading System of Digital Currencies: A Deep Reinforcement Learning with Multidimensional Attention Gating Mechanism. Neurocomputing,volume 402,Pages 171-182,2019 描述 碩士
國立政治大學
經濟學系
108258034資料來源 http://thesis.lib.nccu.edu.tw/record/#G0108258034 資料類型 thesis dc.contributor.advisor 蔡炎龍<br>蕭明福 zh_TW dc.contributor.advisor Thai, YenLung<br>Shaw, MingFu en_US dc.contributor.author (Authors) 蘇育正 zh_TW dc.contributor.author (Authors) Su, Yu-Cheng en_US dc.creator (作者) 蘇育正 zh_TW dc.creator (作者) Su, Yu-Cheng en_US dc.date (日期) 2022 en_US dc.date.accessioned 9-Mar-2023 18:23:05 (UTC+8) - dc.date.available 9-Mar-2023 18:23:05 (UTC+8) - dc.date.issued (上傳時間) 9-Mar-2023 18:23:05 (UTC+8) - dc.identifier (Other Identifiers) G0108258034 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/143773 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 經濟學系 zh_TW dc.description (描述) 108258034 zh_TW dc.description.abstract (摘要) 本研究運用深度強化學習建立虛擬貨幣的投資組合，研究標的主要以 2021年 12 月 31 日市值排名前 50 大的虛擬貨幣。研究期間從 2017 年 1 月 3 日至2021 年 12 月 31 日，並主要以五個因子(Factor):開盤價(Open)、最高價(High)、最低價(Low)、收盤價(Close)、成交量(Volume)為輸入資料(Input)，並在一開始先以(1)市值、(2)平均振幅抓取 30 檔虛擬幣組建投資組合，輸入給深度強化學習模型進行訓練，最終發現相較於其他種因子建立的投資組合，平均振幅打造的投資組合表現更好，也比單一持續持有比特幣來的更合適。 zh_TW dc.description.tableofcontents 誌謝..................................................... 1摘要..................................................... 2目錄..................................................... 3表目錄 ................................................... 5圖目錄 ................................................... 6第一章 緒論 .............................................. 7第一節 研究動機 ..................................... 7第二節 研究目的 ..................................... 8第三節 研究架構與問題設定 ........................... 9第二章 文獻回顧 ......................................... 11第一節 虛擬貨幣市場的資產配置 ...................... 11第二節 強化學習在投資上的應用 ...................... 12第三章 研究設計 ......................................... 13第一節 強化學習模型概述 ............................ 13第二節 深度強化學習與模型設定 ...................... 14第四章 資料來源及實證結果 ............................... 21第一節 資料來源與說明 .............................. 21第二節 模型設定 .................................... 22第三節 實證結果 .................................... 23第五章 結論與未來展望 ................................... 28參考文獻 ................................................ 29 zh_TW dc.format.extent 1600703 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0108258034 en_US dc.subject (關鍵詞) 深度學習 zh_TW dc.subject (關鍵詞) 強化學習 zh_TW dc.subject (關鍵詞) 深度強化學習 zh_TW dc.subject (關鍵詞) 虛擬貨幣 zh_TW dc.subject (關鍵詞) 投資組合 zh_TW dc.subject (關鍵詞) Reinforcement Learning en_US dc.subject (關鍵詞) Portfolio en_US dc.subject (關鍵詞) Crypto en_US dc.title (題名) 運用深度強化學習建立虛擬貨幣投資組合 zh_TW dc.title (題名) Establish The Portfolio of Crypto Currency by Applying Deep-Reinforcement Learning en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) [1] Fan Fang, Carmine Ventre, Michail Basios, Leslie Kanthan, DavidMartinez-Rego, Fan Wu and Lingbo Li. Cryptocurrency trading: a comprehensive survey. Finanical Innovation,8(13),2022.[2] Timothy King and Dimitrios Koutmos. Herding and feedback trading in cryptocurrency markets. Annals of Operations Research,300:79-97,2021.[3] WeiSun, Alisher Tohirovich, Dedahanov, Ho YoungShin and Wei PingLi. Factors affecting institutional investors to add cryptocurrency to asset portfolios. The North American Journal of Economics and Finance,volume 58,2021.[4] Paraskevi Katsiampa, Larisa Yarovaya and DamianZięba. Highfrequency connectedness between Bitcoin and other top-traded crypto assets during the COVID-19 crisis. Journal of International Financial Markets, Institutions and Money,volume 79,2022.[5] Andrés Arévalo, Jaime Niño, G. Hernández and Javier Sandoval. High-Frequency Trading Strategy Based on Deep Neural Networks.International Conference on intelligent Computing, LNAI,volume 9773,2016.[6] Maria Čuljak,BojanTomić and SašaŽiković . Benefits of sectoral cryptocurrency portfolio optimization. Research in International Business and Finance,volume 60,2022.[7] Golnoosh Babaei,Paolo Giudici and EmanuelaRaffinetti. Explainable artificial intelligence for crypto asset allocation. Finance Research Letters,volume 47,Part B,2022.[8] Leonardo Kanashiro Felizardo,Francisco CaioLima Paiva,Catharinede Vita Graves,Elia Yathie Matsumoto,Anna Helena Reali Costa,Emilio DelMoral-Hernandez and Paolo Brandimarte. Outperforming algorithmic trading reinforcement learning systems: A supervised approach to the 30cryptocurrency market. Expert Systems with Applications,volume 202,2022.[10] Hongfeng Xu,Lei Chai,Zhiming Luo and Shaozi Li. Stock movement prediction via gated recurrent unit network based on reinforcement learning with incorporated attention mechanisms. Neurocomputing,volume 467,Pages 214-228,2022.[11] Fengrui Liu,Yang Li,Baitong Li,Jiaxin Li and Huiyang Xie . Bitcoin transaction strategy construction based on deep reinforcement learning. Applied Soft Computing,volume113,Part B,2021.[12] Thibaut Théate and Damien Ernst . An application of deep reinforcement learning to algorithmic trading. Expert Systems with Applications,volume 173,2021.[13] Liguo Weng,Xudong Sun,Min Xia,Jia Liu and Yiqing Xu. Portfolio Trading System of Digital Currencies: A Deep Reinforcement Learning with Multidimensional Attention Gating Mechanism. Neurocomputing,volume 402,Pages 171-182,2019 zh_TW
