股價資料通常為非線性的時間序列，因此準確預測股價一直是專家學者們的目標，鮮少有研究著墨於訓練模型是否能對未知資料進行有效預測。故本研究目的以深度學習模型預測元大台灣50(0050)之淨值，且將預測結果建構實務投資策略，以驗證長短期記憶神經網路(Long Short-Term Memory，簡稱LSTM)對於股價漲跌幅預測之成效，並探討所建構之模型對於不同標的之泛化性(Generalization)。實驗結果顯示，LSTM模型對於所挑選之54檔標的隔日漲跌幅度皆能有效預測，其預測結果所組成之元大台灣50隔日淨值與實際淨值相比，平均誤差為1.08%、標準差為1.07%及RMSE為1.35。最後本研究運用所預測之結果建構投資策略，採用元大台灣50個股期貨進行當沖順勢交易策略，結果顯示績效獲得305.7%之報酬，代表LSTM模型之預測結果，確實能實際運用於交易策略。

Due to nonlinearity and high volatility of financial time series, predict stock market is one of the most important financial time series forecasting problems for both investors and researchers. Some researchers in this field try to perform this problem by proposing different methods and analyses. This research presents a hybrid framework of deep learning model and investment strategy to predict stock market for Yuanta Taiwan Top 50 ETF (0050), which assigns different stock and had compared with the trade return. Experimental results show that the performance is rewarded by 305.7% and the result of the average error is 1.08%, the standard deviation is 1.07%, and RMSE is 1.35, which means that the forecasted results of the LSTM model can indeed be applied to trading strategies.

摘要 ................................ .............................. i
誌 謝 ................................ ........................... iii
目錄 ................................ ............................. iv
圖目錄 ................................ ............................ v
表目錄 ................................ ........................... vi
方程式 ................................ .......................... vii
附錄 ................................ ........................... viii
壹、 緒論 ................................ ...................... 1
一、 研究背景與動機 ................................ ............. 1
二、 研究目的 ................................ ................... 2
三、 研究架構 ................................ ................... 3
貳、 文獻探討 ................................ .................. 4
一、 股價預測模型 ................................ ............... 4
二、 泛化性驗證 ................................ ................. 7
三、 深度學習之績效評估 ................................ ......... 9
參、 研究方法 ................................ ................. 10
一、 研究流程 ................................ .................. 10
二、 研究樣本 ................................ .................. 11
三、 資料收集 ................................ .................. 12
四、 資料前處理 ................................ ................ 13
五、 長短期記憶神經網路模型建構 ................................ 20
六、 交易策略 ................................ .................. 21
肆、 研究結果 ................................ ................. 23
一、 預測目標評估 ................................ .............. 23
二、 泛化性驗證評估 ................................ ............ 24
三、 穩健性實證 (Robust test) ................................ ... 30
四、 績效評估 ................................ .................. 32
伍、 結論與建議 結論與建議 ................................ ............... 38
一、 結論 ................................ ...................... 38
二、 研究貢獻 ................................ .................. 38
三、 後續研究與建議 ................................ ............ 39
參考文獻 ................................ ......................... 41
附錄 ................................ ............................. 46

英文文獻
1.Abbasimehr, H., Shabani, M., & Yousefi, M. (2020) “An optimized model using LSTM network for demand forecasting.”, Computers & industrial engineering, Vol.143, 106435.
2.Balaji, A. J., Ram, D. H., & Nair, B. B. (2018) “Applicability of deep learning models for stock price forecasting an empirical study on BANKEX data.”, Procedia computer science, Vol.143, pp.947-953.
3.Bustos, O., & Pomares-Quimbaya, A. (2020) “Stock market movement forecast: A Systematic review.”, Expert Systems with Applications, Vol.156,113464.
4.Chen, C., Zhang, P., Liu, Y., & Liu, J. (2020) “Financial quantitative investment using convolutional neural network and deep learning technology.”, Neurocomputing, Vol.390, pp.384-390.
5.Chang, P. C., Liu, C. H., Lin, J. L., Fan, C. Y., & Ng, C. S. (2009) “A neural network with a case based dynamic window for stock trading predictio-n.”, Expert Systems with Applications, Vol.36(3), pp.6889-6898.
6.de Oliveira Carosia, A. E., Coelho, G. P., & da Silva, A. E. A. (2021) “Investment Strategies Applied to the Brazilian Stock Market: A Methodology Based on Sentiment Analysis with Deep Learning.”, Expert Systems with Applications, 115470.
7.deAndrés-Galiana, E. J., Bea, G., Fernández-Martínez, J. L., & Saligan, L. N. (2019) “Analysis of defective pathways and drug repositioning in Multiple Sclerosis via machine learning approaches.”, Computers in biology and medicine, Vol.115, 103492.
8.Haq, A. U., Zeb, A., Lei, Z., & Zhang, D. (2021) “Forecasting daily stock tr-end using multi-filter feature selection and deep learning.”, Expert Systems with Applications, Vol.168, 114444.
9.Hiransha, M., Gopalakrishnan, E. A., Menon, V. K., & Soman, K. P. (2018) “NSE stock market prediction using deep-learning models.”, Procedia computer science, Vol.132, pp.1351-1362.
10.Hochreiter, S., & Schmidhuber, J. (1997) “Long short-term memory.”, Neural computation, Vol.9(8), pp.1735-1780.
11.Jiang, M., Liu, J., Zhang, L., & Liu, C. (2020) “An improved Stacking framework for stock index prediction by leveraging tree-based ensemble models and deep learning algorithms.”, Physica A: Statistical Mechanics and its Applications, Vol.541, 122272.
12.Koratamaddi, P., Wadhwani, K., Gupta, M., & Sanjeevi, S. G. (2021) “Market sentiment-aware deep reinforcement learning approach for stock portfolio a-llocation.”, Engineering Science and Technology, an International Journal, Vol.24(4), pp.848-859.
13.Liu, H., & Long, Z. (2020) “An improved deep learning model for predicting stock market price time series.”, Digital Signal Processing, Vol.102,102741.
14.Lei, K., Zhang, B., Li, Y., Yang, M., & Shen, Y. (2020) “Time-driven feature-aware jointly deep reinforcement learning for financial signal representation and algorithmic trading.”, Expert Systems with Applications, Vol.140, 112872.
15.Liu, Q., Tao, Z., Tse, Y., & Wang, C. (2021) “Stock market prediction with deep learning: The case of China.”, Finance Research Letters, 102209.
16.Long, W., Lu, Z., & Cui, L. (2019) “Deep learning-based feature engineering for stock price movement prediction.”, Knowledge-Based Systems, Vol.164, pp.163-173.
17.Li, X., Wu, P., & Wang, W. (2020) “Incorporating stock prices and news sentiments for stock market prediction: A case of Hong Kong.”, Information P-rocessing & Management, Vol.57(5), 102212.
18.Li, X., Shang, W., & Wang, S. (2019) “Text-based crude oil price forecasting: A deep learning approach.”, International Journal of Forecasting, Vol.35(4), pp.1548-1560.
19.Lin, Y., Yan, Y., Xu, J., Liao, Y., & Ma, F. (2021) “Forecasting stock index price using the CEEMDAN-LSTM model.”, The North American Journal of Economics and Finance, Vol.57, 101421.
20.Moghar, A., & Hamiche, M. (2020) “Stock market prediction using LSTM rec-urrent neural network.”, Procedia Computer Science, Vol.170, pp.1168-1173.
21.Moews, B., & Ibikunle, G. (2020) “Predictive intraday correlations in stable and volatile market environments: Evidence from deep learning.”, Physica A: Statistical Mechanics and its Applications, Vol.547, 124392.
22.Mu, W., Rahaman, M., Rios, F. L., Odqvist, J., & Hedström, P. (2021) “Predicting strain-induced martensite in austenitic steels by combining physical modelling and machine learning.”, Materials & Design, Vol.197, 109199.
23.Nevasalmi, L. (2020) “Forecasting multinomial stock returns using machine learning methods.”, The Journal of Finance and Data Science, Vol.6,pp.86-106.
24.Ozbayoglu, A. M., Gudelek, M. U., & Sezer, O. B. (2020) “Deep learning for financial applications: A survey.”, Applied Soft Computing, Vol.93,106384.
25.Qiao, G., Teng, Y., Li, W., & Liu, W. (2019) “Improving volatility forecasting based on Chinese volatility index information: Evidence from CSI 300 index and futures markets.”, The North American Journal of Economics and Finance, Vol.49, pp.133-151.
26.Orimoloye, L. O., Sung, M. C., Ma, T., & Johnson, J. E. (2020) “Comparing the effectiveness of deep feedforward neural networks and shallow architectures for predicting stock price indices.”, Expert Systems with Applications,Vol.139, 112828.
27.Peng, Y., Albuquerque, P. H. M., Kimura, H., & Saavedra, C. A. P. B. (2021) “Feature selection and deep neural networks for stock price direction forecasting using technical analysis indicators.”, Machine Learning with Applica-tions, 100060.
28.Ribeiro, F., & Gradvohl, A. L. S. (2021) “Machine learning techniques applied to solar flares forecasting.”, Astronomy and Computing, Vol.35, 100468.
29.Sharma, D. K., Tokas, B., & Adlakha, L. (2020) “Deep learning in big data and data mining.”, Trends in Deep Learning Methodologies: Algorithms, Applications, and Systems, Vol.37.
30.Shynkevich, Y., McGinnity, T. M., Coleman, S. A., Belatreche, A., & Li, Y. (2017) “Forecasting price movements using technical indicators: Investigating the impact of varying input window length.”, Neurocomputing, Vol.264, pp.71-88.
31.Shen, Z., Zhang, Y., Lu, J., Xu, J., & Xiao, G. (2020) “A novel time series forecasting model with deep learning.”, Neurocomputing, Vol.396,pp.302-313.
32.Vijh, M., Chandola, D., Tikkiwal, V. A., & Kumar, A. (2020) “Stock closing price prediction using machine learning techniques.”, Procedia Computer S-cience, Vol.167, pp.599-606.
33.Wang, Y., Liu, L., & Wu, C. (2020) “Forecasting commodity prices out-of-sample: Can technical indicators help?.”, International Journal of Forecasting, Vol.36(2), pp.666-683.
34.Xie, M., Li, H., & Zhao, Y. (2020) “Blockchain financial investment based on deep learning network algorithm.”, Journal of Computational and Applied Mathematics, Vol.372, 112723.
35.Yin, L., & Yang, Q. (2016) “Predicting the oil prices: do technical indicators help?.”, Energy Economics, Vol.56, pp.338-350.
36.Zhang, X., & Zhang, S. (2021) “Optimal time-varying tail risk network with a rolling window approach.”, Physica A: Statistical Mechanics and its Appli-cations, 126127.
37.Zhang, Y., Ma, F., & Liao, Y. (2020) “Forecasting global equity market volatilities.”, International Journal of Forecasting, Vol.36(4), pp.1454-1475.
中文文獻
1.林儒霆 (2008) 以多項技術指標建構股市投資決策支援系統之研究，中國文化大學資訊管理研究所，碩士論文。
2.陳嘉琪 (2016) 新穎權重調整方式之類神經應用於股價預測問題，國立暨南大學資訊工程學系，碩士論文
3.陳郁夫 (2018)。運用多模式深度強化學習與語義影響模型於交易策略，國立交通大學資訊管理與財務金融學系資訊管理碩士班，碩士論文。
4.林逸青 (2019)。以深度學習建構股價預測模型:以台灣股票市場為例，國立臺中科技大學財務金融研究所碩士班，碩士論文。
5.邱彥誠 (2020)。應用人工智慧於股市新聞與情感分析預測股價走勢，國立台北大學資訊管理研究所，碩士論文。
6.孫永鴻 (2020)。機器學習演算法在股票交易策略之應用。國立臺北大學統計學系碩士，碩士論文
7.陳彥安（2019）。深度學習於台股動態投資組合之應用。國立交通大學管理學院財務金融學程，碩士論文