研究生: |
張奕崴 Yi-Wei Chang |
---|---|
論文名稱: |
結合股票選擇權與股票存託憑證之深度學習股價預測方法 Deep Learning Stock Price Prediction Method Combining Stock Options and Depositary Receipts |
指導教授: |
呂永和
Yung-Ho Leu |
口試委員: |
楊維寧
Wei-Ning Yang 陳雲岫 Yun-Shiow Chen |
學位類別: |
碩士 Master |
系所名稱: |
管理學院 - 資訊管理系 Department of Information Management |
論文出版年: | 2021 |
畢業學年度: | 109 |
語文別: | 中文 |
論文頁數: | 40 |
中文關鍵詞: | 長短期記憶神經網路 、股價預測 、選擇權 |
外文關鍵詞: | LSTM, Stock trend prediction, Options |
相關次數: | 點閱:398 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
隨著網際網路的快速發展,人們可以很方便地進行股市交易,股票市場也被視為最重要的金融市場之一。現今資訊技術也開始應用在傳統金融產業中,並結合深度學習探討市場與股價的關係。然而目前研究方法大多選擇單一股價或是新聞做為參考資料,在股市交易中,選擇權(Options)與存託憑證也是很重要的投資參考選項,其中包含買權(call)與賣權(put)等多項數據。本研究將選擇權具有的多變量與時間序列特徵,與多變量長短期記憶(Multivariate LSTM)模型結合,使模型能有效的學習時間序列中的資訊,並得到最終預測結果。根據實驗結果發現具有多變量的LSTM模型能提升預測股價變化研究的準確度。
With the rapid development of the Internet, people can easily trade in the stock market, and the stock market is also regarded as one of the most important financial markets. In these days, information technology has also begun to be applied in the traditional financial industry, and combined with deep learning to explore the relationship between the market and stock prices. However, most of the current research methods choose a single stock price or news as reference materials. In stock market trading, options and depositary receipts are also very important investment reference options, which include multiple data such as call and put. In this study, the options with time series features are combined with the Multivariate LSTM model to enable the model more effectively learning the information in the time series and obtain the final prediction result. According to the experimental findings, it is found that the multivariate LSTM model can improve the accuracy of the research on predicting stock price changes.
[1]許繼元 (2018年10月)。什麼是選擇權(期權)?如何買賣?Mr. Market市場先生。Retrieved from :
https://rich01.com/what-is-options/
[2]元大期貨。選擇權入門,基本認識。Retrieved from : https://www.yuantafutures.com.tw/futureinvest_01
[3]Chan. K, Li. G, Lin. T. (2013). Informational Content of Options Trading on Acquirer Announcement Return. Journal of Financial and Quantitative Analysis, forthcoming.
[4]蘇斯妤 (2019)。資訊需求與股價聯動關係:以台灣、香港公司之美國存托憑證為例。國立交通大學財務金融研究所。
[5]周佳志 (2017)。具適應性停止策略之深度強化學習法於時間序列早期分類之研究。國立成功大學工業與資訊管理學系。
[6]陳俊宏 (2015)。應用線上機器學習演算法於財務時間序列預測問題 — 以美國S&P500成分股為例。臺北市立大學資訊科學系。
[7]林楓淳 (2006)。結合ARIMA與支援向量迴歸於財務時間序列預測模式之建構-以新加坡交易所日經225指數期貨為例。輔仁大學管理學研究所。
[8]張永承 (2011)。應用類神經網路方法於金融時間序列預測之研究-以TWSE台股指數為例。國立政治大學資訊管理研究所。
[9]A. A. Ariyo, A. O. Adewumi, C. K. Ayo. (2014), Stock Price Prediction Using the ARIMA Model, 2014 UKSim-AMSS 16th International Conference on Computer Modelling and Simulation.
[10]D. Reddy. (2019). Stock price prediction and forecast using LSTM and ARIMA. Retrieved from :
https://medium.com/@dhirajreddy13/stock-price-prediction-and-forecast-using-lstm-and-arima-52db753a23c7
[11]M. Paradkar. (2017). Forecasting Stock Returns Using ARIMA Model. Retrieved from :
https://blog.quantinsti.com/forecasting-stock-returns-using-arima-model/
[12]J. Chung, C. Gulcehre, K. Cho.(2014). Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. Cornell University.
[13]A. Gomez. (2016). Backpropogating an LSTM: A Numerical Example. Retrieved from :
https://medium.com/@aidangomez/let-s-do-this-f9b699de31d9
[14]郭文秀 (2014)。市場狀態和台灣選擇權交易活動與股價關係之探討。應用經濟論叢,第九十五期,頁101-145。
[15]Jason Brownlee (2017). Deep Learning for Time Series Forecasting. Machine Learning Mastery Pty.