跳到主要內容

臺灣博碩士論文加值系統

(44.212.99.208) 您好!臺灣時間:2024/04/17 21:05
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:陳郁蓁
研究生(外文):CHEN,YU-CHEN
論文名稱:利用卷積神經網絡建構股票價格預測模型:以標準普爾 500 指數為例
論文名稱(外文):Constructing stock price forecast model with Convolutional Neural Networks: A Case of S&P500 Index
指導教授:黃文楨黃文楨引用關係
指導教授(外文):HUANG,WEN-CHEN
口試委員:殷堂凱陳彥銘黃文楨
口試委員(外文):YIN,TANG-KAICHEN,YEN-MINGHUANG,WEN-CHEN
口試日期:2020-06-12
學位類別:碩士
校院名稱:國立高雄科技大學
系所名稱:資訊管理系
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:中文
論文頁數:67
中文關鍵詞:股價預測深度學習卷積神經網路長短期記憶模型貝葉斯優化
外文關鍵詞:Stock price forecastDeep learningConvolutional neural networksLong short - term memoryBayesian optimization
相關次數:
  • 被引用被引用:1
  • 點閱點閱:367
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
根據行政院主計總處國勢普查處對於 2018 年台灣平均每人資產分佈的報告 指出,台灣人有高達 15.46%的資產是分佈在股票市場上,顯示出了股票在打造 被動資產時的重要性。近年來也有許多學者利用人工智慧來預測股價趨勢,其特 徵資料多數僅使用金融指標,最終也獲得了不錯的報酬率。
本研究目的在於探討金融指標與石油價格、黃金價格、黃金價格波動指數和 石油價格波動指數等特徵資料,在 S&P500 和個股間的影響性,以及對於卷積神 經網路(CNN)和長短期記憶模型(LSTM)的適用性。研究方法採用二分類(上 漲、下跌)與三分類(買進、賣出、持有)的目標資料標記方法做預測。最後針 對研究結果做分析,並使用貝葉斯優化超參數與更新分類權重等方式,提升模型 準確性和投資報酬率(ROI)。
研究結果顯示黃金價格、黃金波動指數、原油價格、原油價格波動指數是可 以做為模型輸入特徵的。在二分類的情況下,可以預測未來 10 日後的股票價格 將比現在上漲或下跌,模型正確率達到 67%。CNN 和 LSTM 在此分類情況下不分 軒輊;而三分類時,該類型特徵與金融指標結合,作為模型輸入特徵用於卷積神 經網路時,可以找到較佳的買點和賣點,比長短期記憶模型的分類結果更為準確。 針對樣本外 2019 年預測的買點與賣點,實測後可達到 13.23%的 ROI。最後,在 個股的投資報酬率回測中,發現在半導體與能源相關產業時參考黃金價格、黃金 波動指數、原油價格、原油價格波動指數會使其投資報酬率更佳。而服飾與食品 類則不適用此類特徵。
According to the report of the National Survey Office of the Executive Yuan's General Statistics Office on the distribution of Taiwan's average assets per person in 2018, as many as 15.46% of Taiwanese's assets are distributed on the stock market, showing that the importance of stocks in creating passive assets. In recent years, many scholars have also used artificial intelligence to predict stock price trends. Most of their characteristic data only use financial indicators, and they have finally obtained a good rate of return.
The purpose of this research is to explore the financial indicators and the characteristic data such as oil price, gold price, gold price volatility index and oil price volatility index, the influence between S&P500 and individual stocks, as well as the suitability for the convolutional neural network (CNN) and long-term and short-term memory model (LSTM). The research method uses the target data labeling method of two categories (up, down) and three categories (buy, sell, hold) to make predictions. Finally, we analyze the research results, use Bayesian optimization hyperparameters and update classification weights to improve model accuracy and increase return on investment (ROI).
The research results show that gold price, gold volatility index, crude oil price, and crude oil price volatility index can be used as input features of the model. In the case of the second classification, it can be predicted that the stock price will rise or fall in the next 10 days from the present, and the model accuracy rate reaches 67%. In this classification, CNN and LSTM do not distinguish dramatically; in the three classification, this type of feature is combined with financial indicators, and when used as a model input feature for CNN, better buying and selling points can be found, which is better than LSTM. The classification results of the memory model are more accurate. For the buying points and selling points predicted of test data, the measured ROI can reach 13.23%. Finally, in the back-test of individual stocks’ return on investment, it was found that referring to the price of gold, gold volatility index, crude oil price, and crude oil price volatility index in the semiconductor and energy-related industries would make their investment returns better. Apparel and food categories do not apply this feature.
中文摘要
英文摘要
誌謝
目錄
表目錄
圖目錄
壹、緒論
一、研究背景
二、研究動機
三、研究目的
四、研究貢獻與重要性
五、研究架構
貳、文獻探討
一、深度學習起源與應用
二、長短期記憶模型
三、卷積神經網路
四、超參數優化
五、貝葉斯優化
六、人工智慧應用於股價預測相關文獻
七、結語
參、研究方法與設計
一、預測目標
二、股價指數與原油和黃金的關係
(一)股票指數與原油價格
(二)股票指數與期貨市場
(三)黃金價格與原油價格
(四)黃金價格與股票指數
三、分類方式
(一)二分類標記方式
(二)三分類標記方式
四、資料集
五、模型輸入特徵
六、模型參數設計與優化
(一)深度學習模型的模型參數
1. 模型設計變量
2. 超參數
(二)模型設計變量的參數選擇
1. 優化器
2. 損失函數
3. 激發函數
(三)超參數的選擇
七、本研究所用之模型架構與參數
(一)CNN 模型架構說明
(二)LSTM 模型架構說明
八、三分類權重優化方式
九、模型評估方式
十、投資報酬率算法
(一)報酬率計算公式
(二)交易程式碼說明
肆、實驗結果與分析
一、二分類實驗結果
(一)測試集評估特徵有效性
(二)測試集評估模型適用性
(三)Bayesian Optimization 優化模型超參數
(四)S&P 500 測試集預測準確率與時間關係
二、三分類實驗結果
(一)三分類標記方式探討
(二)測試集評估特徵正確率和報酬率
(三)測試集評估模型有效性與報酬率
(四)分次加碼買進投資比較
(五)相關文獻結果比較
三、個股回測報酬率
(一)S&P 500 中在 2019 年表現較佳的 9 間公司
(二)S&P 500 中在 2019 年表現較差的 9 間公司
伍、研究結論與建議
一、研究結論
(一)二分類研究結論
(二)三分類研究結論
二、研究建議
參考文獻
1. Bergstra, J., & Bengio, Y. (2012). Random search for hyper-parameter optimization. Journal of machine learning research, 13(Feb), 281-305.
2. Choi, H. K. (2018). Stock price correlation coefficient prediction with ARIMA- LSTM hybrid model. arXiv preprint arXiv:1808.01560.
3. Frazier, P. I. (2018). A tutorial on Bayesian optimization. arXiv preprint arXiv:1807.02811.
4. Fu, X., Du, J., Guo, Y., Liu, M., Dong, T., & Duan, X. (2018). A Machine Learning Framework for Stock Selection. arXiv preprint arXiv:1806.01743.
5. Gokmenoglu, K. K., & Fazlollahi, N. (2015). The interactions among gold, oil, and stock market: evidence from S&P500. Procedia Economics and Finance, 25(Supp. C), 478-488.
6. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
7. Hsu, C. M. (2011). A hybrid procedure for stock price prediction by integrating self-organizing map and genetic programming. Expert Systems with Applications, 38(11), 14026-14036.
8. Huang, B., Ding, Q., Sun, G., & Li, H. (2018, February). Stock Prediction based on Bayesian-LSTM. In Proceedings of the 2018 10th International Conference on Machine Learning and Computing (pp. 128-133).
9. Kingma, D. P., Salimans, T., & Welling, M. (2015). Variational dropout and the local reparameterization trick. In Advances in neural information processing systems (pp. 2575-2583).
10. Kouassi, K., & Moodley, D. (2019). Automatic construction of neural networks for stock market trend prediction.
11. Lai, C. Y., Chen, R. C., & Caraka, R. E. (2019, July). Prediction Stock Price Based on Different Index Factors Using LSTM. In 2019 International Conference on Machine Learning and Cybernetics (ICMLC) (pp. 1-6). IEEE.
12. Lakshminarayanan, S. K., & McCrae, J. (2019). A Comparative Study of SVM and LSTM Deep Learning Algorithms for Stock Market Prediction.
13. Lee, S. I. (2020). Hyperparameter Optimization for Forecasting Stock Returns. arXiv preprint arXiv:2001.10278.
14. Li, H., Shen, Y., & Zhu, Y. (2018, November). Stock price prediction using attention-based multi-input LSTM. In Asian Conference on Machine Learning (pp. 454-469).
15. Liu, J., Chao, F., Lin, Y. C., & Lin, C. M. (2019). Stock Prices Prediction using Deep Learning Models. arXiv preprint arXiv:1909.12227.
16. Malkiel, B. G. (1999). A random walk down Wall Street: including a life-cycle guide to personal investing. WW Norton & Company
17. Nelson, D. M., Pereira, A. C., & de Oliveira, R. A. (2017, May). Stock market's price movement prediction with LSTM neural networks. In 2017 International joint conference on neural networks (IJCNN) (pp. 1419-1426). IEEE.
18. Patel, J., Shah, S., Thakkar, P., & Kotecha, K. (2015). Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques. Expert systems with applications, 42(1), 259-268.
19. Sezer, O. B., & Ozbayoglu, A. M. (2018). Algorithmic financial trading with deep convolutional neural networks: Time series to image conversion approach. Applied Soft Computing, 70, 525-538.
20. Shravya Shetty, M.S. & Daniel Tse, M.D. (2020). Using AI to improve breast cancer screening
21. Sim, H. S., Kim, H. I., & Ahn, J. J. (2019). Is deep learning for image recognition applicable to stock market prediction? Complexity, 2019.
22. Siripurapu, A. (2014). Convolutional networks for stock trading.
23. Smith, G. (2001). The price of gold and stock price indices for the United States. The World Gold Council, 8(1), 1-16.
24. Sun, Y., Wang, X., & Tang, X. (2014). Deep learning face representation from predicting 10,000 classes. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 1891-1898). Ticknor, J. L. (2013). A Bayesian regularized artificial neural network for stock market forecasting. Expert Systems with Applications, 40(14), 5501-5506.
25. Vanstone, B., Finnie, G., & Hahn, T. (2012). Creating trading systems with fundamental variables and neural networks: The Aby case study. Mathematics and computers in simulation, 86, 78-91.
26. Wu, N., Phang, J., Park, J., Shen, Y., Huang, Z., Zorin, M., ... & Wolfson, S. (2019). Deep neural networks improve radiologists’ performance in breast cancer screening. IEEE transactions on medical imaging.
27. Yuan, X., Yuan, J., Jiang, T., & Ain, Q. U. (2020). Integrated Long-Term Stock Selection Models Based on Feature Selection and Machine Learning Algorithms for China Stock Market. IEEE Access, 8, 22672-22685.
28. Zhang, K., Zhong, G., Dong, J., Wang, S., & Wang, Y. (2019). Stock market prediction based on generative adversarial network. Procedia computer science, 147, 400-406.
29. Zhu, L., & Laptev, N. (2017, November). Deep and confident prediction for time series at uber. In 2017 IEEE International Conference on Data Mining Workshops (ICDMW) (pp. 103-110). IEEE.
30. 薛淑如,(2003),應用類神經網路於股票價值之評估,國立彰化師範大學會 計學系碩士論文。
31. 楊柳勇、史震濤,(2004),黃金價格的長期決定因素分析,統計研究,2004, 21(6):21-24。
32. 沈沛瑄、魏廉臻、張瑞益,(2019),以 LSTM-RNN 預測 ETF 50 股價趨勢 並結合交易策略以獲取最大獲利率,2019 全國計算機會議。
電子全文 電子全文(網際網路公開日期:20250607)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊