Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/88779Full metadata record
| ???org.dspace.app.webui.jsptag.ItemTag.dcfield??? | Value | Language |
|---|---|---|
| dc.contributor.advisor | 陳永耀 | zh_TW |
| dc.contributor.advisor | Yung-Yaw Chen | en |
| dc.contributor.author | 楊明泰 | zh_TW |
| dc.contributor.author | Ming-Tai Yang | en |
| dc.date.accessioned | 2023-08-15T17:45:10Z | - |
| dc.date.available | 2023-11-09 | - |
| dc.date.copyright | 2023-08-15 | - |
| dc.date.issued | 2022 | - |
| dc.date.submitted | 2023-08-07 | - |
| dc.identifier.citation | <BIS Triennial Central Bank Survey, 2022>.
Cavalcante, R.C., et al., Computational intelligence and financial markets: A survey and future directions. Expert Systems with Applications, 2016. 55: p. 194-211. Galeshchuk, S. and S. Mukherjee, Deep networks for predicting direction of change in foreign exchange rates. Intelligent Systems in Accounting, Finance and Management, 2017. 24(4): p. 100-110. Nelson, D.M., A.C. Pereira, and R.A. De Oliveira. Stock market's price movement prediction with LSTM neural networks. in 2017 International joint conference on neural networks (IJCNN). 2017. Ieee. Chourmouziadis, K. and P.D. Chatzoglou, An intelligent short term stock trading fuzzy system for assisting investors in portfolio management. Expert Systems with Applications, 2016. 43: p. 298-311. Zhang, K., et al., Stock market prediction based on generative adversarial network. Procedia computer science, 2019. 147: p. 400-406. Sezer, O.B. and A.M. Ozbayoglu, Algorithmic financial trading with deep convolutional neural networks: Time series to image conversion approach. Applied Soft Computing, 2018. 70: p. 525-538. Patel, J., et al., Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques. Expert systems with applications, 2015. 42(1): p. 259-268. Maqsood, H., et al., A local and global event sentiment based efficient stock exchange forecasting using deep learning. International Journal of Information Management, 2020. 50: p. 432-451. Wen, M., et al., Stock market trend prediction using high-order information of time series. Ieee Access, 2019. 7: p. 28299-28308. Chen, J.-F., et al. Financial time-series data analysis using deep convolutional neural networks. in 2016 7th International conference on cloud computing and big data (CCBD). 2016. IEEE. Ni, L., et al., Forecasting of forex time series data based on deep learning. Procedia computer science, 2019. 147: p. 647-652. Qi, L., M. Khushi, and J. Poon. Event-driven LSTM for forex price prediction. in 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE). 2020. IEEE. Alonso-Monsalve, S., et al., Convolution on neural networks for high-frequency trend prediction of cryptocurrency exchange rates using technical indicators. Expert Systems with Applications, 2020. 149: p. 113250. Fischer, T. and C. Krauss, Deep learning with long short-term memory networks for financial market predictions. European journal of operational research, 2018. 270(2): p. 654-669. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/88779 | - |
| dc.description.abstract | 根據國際清算銀行調查資料，外匯市場每天平均交易量約為7.2萬億美元，這代表著外匯市場擁有巨額的交易量和獲利潛力。然而，與股票市場相比，外匯市場的價格波動性較低。為了在外匯交易中獲取利潤，短線交易可能更為適合。此外，在交易過程中，買賣信號的定義方式也是重要的研究議題。因此，本研究旨在討論不同的交易信號定義在外匯市場短線交易的可行性，並將其應用於實際交易情境中。
研究中的主要方法是以每分鐘的美元/新台幣資料作為基礎，透過趨勢指標和深度學習方法（CNN）捕捉外匯市場的趨勢和模式，以提供準確的買賣點建議。同時，研究也討論不同買賣點定義方式和考慮實際交易的成本因素，例如報價差，以更準確地評估實際交易情形。透過訓練和評估每分鐘的美元/新台幣資料，並結合交易演算法執行實際的買賣交易，本研究將探討其在短線交易中的潛在應用價值。 在本研究中，交易信號的標籤旨在一定範圍內最大化利潤，同時排除在盤整期間標籤。無論是使用帶有交易成本過濾的DP還是採用標籤過濾和窗口方法，都可以達到這種效果。從短期交易結果來看，使用DP法的利潤比買入持有策略高出7倍，而窗口法的利潤則高出10倍。 | zh_TW |
| dc.description.abstract | According to the survey from BIS, the average daily trading volume in the Forex market is around 7.2 trillion US dollars, indicating the market's significant trading volume and profit potential. However, compared to the stock market, the Forex market has lower price volatility. Short-term trading may be more suitable for making a profit in Forex trading. Additionally, the definition of buy and sell signals is a hot topic in the trading process. Therefore, this study aims to discuss the feasibility of different trade signal definitions in short-term trading in the Forex market and apply them in real trading scenarios.
The primary approach is based on minute-level USD/TWD data, utilizing trend indicators and a 3D CNN model to capture trends and patterns in the Forex market. The research also considers the transaction costs, such as spreads, to more accurately assess the practical trading and discuss different trading signal definitions. By evaluating minute-level USD/TWD data and implementing trading algorithms for actual transactions, this study aims to explore their potential value in short-term trading. In this study, the trading signals is designed to maximize profits within a specific range while excluding labels during consolidation periods. Both can achieve this effect by using DP with transaction cost filtering or employing label filtering and window methods. In short-term trading results, using the DP method can yield profits seven times higher than the B&H strategy, while the window method results in profits ten times higher. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2023-08-15T17:45:10Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2023-08-15T17:45:10Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 口試委員會審定書 #
誌謝 ii 中文摘要 iii ABSTRACT iv CONTENTS v LIST OF FIGURES viii LIST OF TABLES x Chapter 1 Introduction 1 1.1 Background 4 1.1.1 Introduction to the exchange rates 4 1.1.2 Ways to make profit in exchange rates 7 1.1.3 Transaction cost in the Forex market 9 1.2 Problem Statements 11 1.3 Aims of the Thesis 12 1.4 Thesis Structure 14 Chapter 2 Literature Reviews 15 2.1 Price or Price movement Forecasting using Deep Learning 15 2.2 Financial Forecasting using Convolutional Neural Network 18 2.3 Summary of Literature Reviews 24 Chapter 3 Methodology 25 3.1 Workflow 26 3.2 Dataset Information 27 3.3 Feature Map Generation 30 3.4 Trading Signals Labeling Method 32 3.4.1 Optimal entry and exit points based on dynamic programming. 32 3.4.2 Entry and exit points based on window size method identifying local maximum or minimum. 34 3.5 Label Filtering Applying on Window Size Method 36 3.6 Deep Learning Architecture 39 3.6.1 Data preprocessing 39 3.6.2 Data imbalance problem 40 3.6.3 Training process and CNN model architecture 42 3.7 Financial Evaluation Method 45 Chapter 4 Results and Discussion 46 4.1 Experimental Setup 46 4.1.1 Specification of hardware and software 47 4.1.2 Hyperparameter Setting up 48 4.2 DP Labeling Method Results 50 4.2.1 USD/TWD minute data characteristics 50 4.2.2 DP labeling characteristics in consolidation periods 52 4.2.3 Financial performance of DP method 56 4.3 Labeling Filtering Applying on Window Size Method Results 59 4.3.1 Model performance using label filtering 59 4.3.2 Financial performance using label filtering 62 4.4 Experimental Results on Different Window Size 66 4.4.1 Model performance in different window size 68 4.4.2 Financial performance in different window size 73 4.5 Comparison of Return Rate of the Proposed System 77 Chapter 5 Conclusions and Future Work 78 REFERENCE 79 | - |
| dc.language.iso | en | - |
| dc.subject | 交易訊號 | zh_TW |
| dc.subject | 卷積神經網路 | zh_TW |
| dc.subject | 趨勢指標 | zh_TW |
| dc.subject | 外匯市場 | zh_TW |
| dc.subject | 短線交易 | zh_TW |
| dc.subject | 深度學習 | zh_TW |
| dc.subject | short-term trading | en |
| dc.subject | deep learning | en |
| dc.subject | convolutional neural network | en |
| dc.subject | trend indicators | en |
| dc.subject | Forex market | en |
| dc.subject | Trading signals | en |
| dc.title | 以卷積神經網路和趨勢指標建立外匯交易市場之交易策略 | zh_TW |
| dc.title | Trading Strategies Based on Convolutional Neural Network and Trend Indicators in Foreign Exchange Trading | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 111-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 葉倚任;張智星 | zh_TW |
| dc.contributor.oralexamcommittee | Yi-Ren Yeh;Jyh-Shing Jang | en |
| dc.subject.keyword | 外匯市場,短線交易,趨勢指標,深度學習,卷積神經網路,交易訊號, | zh_TW |
| dc.subject.keyword | Forex market,short-term trading,trend indicators,deep learning,convolutional neural network,Trading signals, | en |
| dc.relation.page | 80 | - |
| dc.identifier.doi | 10.6342/NTU202303204 | - |
| dc.rights.note | 未授權 | - |
| dc.date.accepted | 2023-08-09 | - |
| dc.contributor.author-college | 電機資訊學院 | - |
| dc.contributor.author-dept | 電機工程學系 | - |
| Appears in Collections: | 電機工程學系 | |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| ntu-111-2.pdf Restricted Access | 2.94 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.