Please use this identifier to cite or link to this item:
https://ah.lib.nccu.edu.tw/handle/140.119/124936| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | 蔡炎龍<br>蕭明福 | zh_TW |
| dc.contributor.author | 吳凱華 | zh_TW |
| dc.contributor.author | Wu, Kai-Hua | en_US |
| dc.creator | 吳凱華 | zh_TW |
| dc.creator | Wu, Kai-Hua | en_US |
| dc.date | 2019 | en_US |
| dc.date.accessioned | 2019-08-07T08:48:30Z | - |
| dc.date.available | 2019-08-07T08:48:30Z | - |
| dc.date.issued | 2019-08-07T08:48:30Z | - |
| dc.identifier | G0106258009 | en_US |
| dc.identifier.uri | http://nccur.lib.nccu.edu.tw/handle/140.119/124936 | - |
| dc.description | 碩士 | zh_TW |
| dc.description | 國立政治大學 | zh_TW |
| dc.description | 經濟學系 | zh_TW |
| dc.description | 106258009 | zh_TW |
| dc.description.abstract | 交易所買賣基金(Exchange Traded Funds, ETF)有別於個股投資，具有分散風險的特性，是一種追蹤特定股價指數的投資商品，也就是一種將股票指數商品化並長期持有的金融商品。\n持有金融商品的目的就是獲利，因此價格或趨勢的預測準確率就變得相當的重要。文獻上實證發現類神經網路較傳統時間序列方法的預測能力高，加上近年機器學習快速發展，本文以類神經網路長短期記憶模型與生成對抗網路為研究方法，建立一個能廣泛運用在台灣非金融類交易所買賣基金的價格與走勢預測。變數除了有收盤價與成交量之外，交易所買賣基金屬於長期持有的商品，產業與總體的變化也是影響行情走勢的重要因素，因此加入匯豐台灣製造業採購經理人指數做為總體變數。此外，為了捕捉總體變數造成的價格影響，加入二十日與四十五日的收盤價移動平均捕捉價格趨勢。\n實證結果發現，使用長短期記憶模型具有預測波動較大的台灣非金融類交易所買賣基金之收盤價格能力，而生成對抗網路具有較高的預測漲跌能力，且行情確實為牛市的時候，生成對抗網路也有較高的能力夠捕捉此趨勢。 | zh_TW |
| dc.description.tableofcontents | 致謝詞 I\n摘要 II\n目錄 III\n表目錄 IV\n圖目錄 V\n第一章 緒論 1\n第一節 研究動機 1\n第二節 研究架構 3\n第二章 文獻回顧 4\n第一節 深度學習時間序列資料應用文獻 4\n第二節 深度學習金融價格預測應用文獻 6\n第三章 研究方法 8\n第一節 模型方法 8\n一 長短期記憶模型 8\n二 生成對抗網路 10\n三 卷積神經網路 12\n第二節 活化函數 13\n第三節 損失函數 15\n第四節 梯度下降法 16\n第四章 資料處理 18\n第一節 資料與特徵選取 18\n第二節 訓練方式 20\n第三節 模型參數設定 22\n第五章 實證結果 24\n第一節 長短期記憶模型實證結果 24\n第二節 生成對抗網路實證結果 26\n第三節 各模型結果比較 28\n一 長短期記憶模型與生成對抗網路預測富邦台灣摩根結果 28\n二 長短期記憶模型與生成對抗網路預測富邦台灣采吉50基金結果 35\n第四節 小結 42\n第六章 結論與建議 44\n第一節 結論 44\n第二節 後續研究與建議 45\n參考文獻 46 | zh_TW |
| dc.format.extent | 3519420 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.source.uri | http://thesis.lib.nccu.edu.tw/record/#G0106258009 | en_US |
| dc.subject | 深度學習 | zh_TW |
| dc.subject | 類神經網路 | zh_TW |
| dc.subject | 交易所買賣基金 | zh_TW |
| dc.title | 以深度學習模型預測台灣ETF價格走勢 | zh_TW |
| dc.type | thesis | en_US |
| dc.relation.reference | 中文文獻\n胡依淳(2018)，「深度卷積神經網路中卷積層之分析及比較」，國立暨南國際大學電機工程學系碩士論文。\n\n陳全溢(2018)，「結合類神經網路預測與投資策略於台灣50指數股票型基金之操作」，國立中興大學資訊管理學系碩士論文。\n\n陳俊諺(2018)，「運用類神經網路與田口法預測台灣ETF指數」，中原大學，資訊管理系碩士論文。\n\n黃焜烽(2018)，「利用深度類神經網路模型預測台灣股價指數走勢」，國立臺北大學經濟系碩士論文。\n\n楊國良(2017)，「運用倒傳遞類神經網路預測台灣50指數ETF股價走勢」，國立金門大學理工學院工程科技碩士在職專班資訊系統組碩士論文。\n \n英文文獻\nAdam Santoro, Ryan Faulkner, David Raposo, Jack Rae, Mike Chrzanowski, Théophane Weber, Daan Wierstra, Oriol Vinyals, Razvan Pascanu, Timothy Lillicrap. (2018). Relational recurrent neural networks. NeurIPS 2018.\n\nAlex Graves, Greg Wayne, Ivo Danihelka. (2014). Neural Turing Machines. arXiv:1410.5401v2\n\nAmin Hedayati Moghaddama, Moein Hedayati Moghaddamb, Morteza Esfandyaric. (2016). Stock market index prediction using artificial neural network. Journal of Economics, Finance and Administrative Science Volume 21, Issue 41, 89–93\n\nAshish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, Illia Polosukhin. (2017). Attention Is All You Need. arXiv:1706.03762v5.\n\nDzmitry Bahdanau, Kyunghyun Cho, Yoshua Bengio. (2016). Neural Machine Translation by Jointly Learning to Align and Translate. arXiv:1409.0473v7.\n\nIan J. Goodfellow, Jean Pouget-Abadiey, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozairz, Aaron Courville, Yoshua Bengio. (2014). Generative Adversarial Nets. NIPS 2014.\n\nIlya Sutskever, Oriol Vinyals, Quoc V. Le. (2014). Sequence to Sequence Learning with Neural Networks. NIPS 2014.\n\nJohn Gamboa (2017). Deep Learning for Time-Series Analysis. arXiv:1701.01887v1.\n\nKangZhang, GuoqiangZhong, JunyuDong, ShengkeWang, YongWang. (2019). Stock Market Prediction Based on Generative Adversarial Network. Procedia Computer Science, Volume 147, 2019, Pages 400-406.\n\nKyunghyun Cho, Bart van Merrienboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, Yoshua Bengio. (2014). Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation. arXiv:1406.1078v3.\n\nMasaya Abe1 and Hideki Nakayama2. (2018). Deep Learning for Forecasting Stock Returns in the Cross-Section. arXiv:1801.01777v4.\n\nMikołaj Bi´nkowski, Gautier Marti, Philippe Donnat. (2018). Autoregressive Convolutional Neural Networks for Asynchronous Time Series. arXiv:1703.04122v4.\n\nMinh-Thang Luong Hieu Pham Christopher D. Manning. (2015). Effective Approaches to Attention-based Neural Machine Translation. arXiv:1508.04025v5.\n\nS.E. Yi, A. Viscardi, T. Hollis. (2018). A Comparison of LSTMs and Attention Mechanisms for Forecasting Financial Time Series. arXiv:1812.07699v1\n\nSepp Hochreiter and Jurgen Schmidhuber. (1997). LONG SHORT-TERM MEMORY. Neural Computation 9(8):1735-1780.\n\nSIMA SIAMI NAMIN1, AKBAR SIAMI NAMIN2. (2018). FORECASTING ECONOMIC AND FINANCIAL TIME SERIES: ARIMA VS. LSTM. arXiv:1803.06386v1.\n\nT. Kimoto ; K. Asakawa ; M. Yoda ; M. Takeoka. (1990). Stock market prediction system with modular neural networks.IEEE 10.1109/IJCNN.1990.137535.\n\nTakashi MATSUBARA, Member, Ryo AKITA, Nonmember, and Kuniaki UEHARA. (2018). Stock Price Prediction by Deep Neural Generative Model of News Articles. IEICE Transactions on Information and Systems, 2018 Volume E101.D Issue 4, Pages 901-908.\n\nThomas R. Cook and Aaron Smalter Hall. (2017). Macroeconomic Indicator Forecasting with Deep Neural Networks. Federal Reserve Bank of Kansas City, Research Working Paper 17-11, September 2017\n\nTrieu H. Trinh1 Andrew M. Dai Minh-Thang Luong Quoc V. Le. (2018). Learning Longer-term Dependencies in RNNs with Auxiliary Losses. ICLR 2018.\n\nV.V.Kondratenko, Yu.A Kuperin.(2003).Using Recurrent Neural Networks To Forecasting of Forex. arXiv:cond-mat/0304469v1.\n\nV´ıctor Campos, Brendan Jouz, Xavier Gir´o-i-Nietox, Jordi Torresy, Shih-Fu Chang. (2018) “SKIP RNN: LEARNING TO SKIP STATE UPDATES IN RECURRENT NEURAL NETWORKS. ICLR 2018\n\nWei Bao, Jun Yue, Yulei Rao. (2017). A deep learning framework for financial time series using stacked autoencoders and long-short term memory. PLoS ONE 12(7): e0180944. https://doi.org/10.1371/journal.pone.0180944\n\nXiao Dingy, Yue Zhangz, Ting Liuy, Junwen Duan. (2015). Deep Learning for Event-Driven Stock Prediction. IJCAI 2015.\n\nXingyu Zhou, Zhisong Pan, Guyu Hu, Siqi Tang, Cheng Zhao. (2018). Stock Market Prediction on High-Frequency Data Using Generative Adversarial Nets. Mathematical Problems in Engineering, Volume 2018, Article ID 4907423, 11 pages.\n\nXin-Yao Qian. (2017). Financial Series Prediction: Comparison Between Precision of Time Series Models and Machine Learning Methods. arXiv:1706.00948. | zh_TW |
| dc.identifier.doi | 10.6814/NCCU201900589 | en_US |
| item.fulltext | With Fulltext | - |
| item.openairecristype | http://purl.org/coar/resource_type/c_46ec | - |
| item.grantfulltext | embargo_20240723 | - |
| item.openairetype | thesis | - |
| item.cerifentitytype | Publications | - |
| Appears in Collections: | 學位論文 | |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| 800901.pdf | 3.44 MB | Adobe PDF2 | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.