本文探討了如何應用深度學習技術預測新加坡房地產投資信託基金(Singapore Real Estate Investment Trust, S-REIT)價格，透過引入各種技術來建構模型，並提供對 其預測表現的比較，其中 S-REIT 以產業類別進行區分。在此過程中，本文採用的資 料，不僅包括傳統的金融指標，還包括文字資訊，以討論加入文本分析對預測模型的 增強潛力。本文進一步區分了 COVID-19 疫情期間和疫情解封之後的情景，以評估疫 情大流行對模型效能的影響，在前所未有的全球經濟變化之後提供即時的發現。結果 顯示，醫療保健業、餐旅業、零售業、工業與數據中心業在疫情期間，不同 Document to Vector (Doc2Vec) 及降維方法下，加文字資訊模型均顯著提升，顯示對這些產業而 言，在疫情期間有效的新聞文字資訊挖掘，對提升模型預測效能佔有關鍵的地位;然 而，疫情解封後沒有特定模型組合具有顯著的優勢，特定產業背景下，也未見絕對優 越模型組合。另外，實驗顯示雙向長短期記憶網絡(Bidirectional Long Short-Term Memory, Bi-LSTM)的時間序列分析能力在實際應用中未穩定超越長短期記憶網絡 (Long Short-Term Memory, LSTM)模型，表明模型複雜性不一定等同更佳預測效 能，強調未來研究應考慮實際資料特性及預測需求，避免盲目追求結構複雜化。

This article explores how to apply deep learning technology to predict the price of Singapore Real Estate Investment Trust (S-REIT). It introduces various technologies to construct the model and provides a comparison of its prediction performance. Among them, S-REIT are distinguished by sector. In this process, the data used in this article include not only traditional financial indicators, but also textual information to discuss the potential of adding textual analysis to enhance predictive models. This article further distinguishes between scenarios during and after the epidemic in Singapore to assess the impact of the pandemic on model performance and provide immediate findings in the wake of unprecedented global economic changes. The results show that during the epidemic period in sectors of Healthcare, Hospitality, Retail, and Industrial & Data Centre, models are significantly improved under different Document to Vector (Doc2Vec) and dimensionality reduction methods on textual data. This proves that effective news text mining during the epidemic plays a key role in improving model prediction performance; however, no specific model combination performs significantly better after the epidemic is over, and none of these is consistently superior among different sectors. In addition, experiments show that the time series analysis capability of the Bidirectional Long Short-Term Memory (Bi-LSTM) model has not stably surpassed the Long Short-Term Memory (LSTM) model in practical applications, indicating that the model complexity does not necessarily equate to better prediction performance. It is emphasized that future research should consider actual data characteristics and prediction needs, and that people should avoid blindly pursuing structural complexity.

摘要 i
目錄 iii
表目錄 v
圖目錄 vi
第一章 緒論 1
第一節 研究背景 1
第二節 研究動機 2
第二章 市場背景介紹和文獻回顧 3
第一節 市場背景介紹 3
第二節 文獻回顧 4
一、 時間序列與機器學習 4
二、 文本分析 6
三、 REIT 7
四、 小結 8
第三章 資料 10
第一節 資料介紹 10
第二節 技術指標 13
第四章 模型方法 14
第一節 分詞和文本特徵提取：Document to Vector (Doc2Vec) 15
第二節 文字特徵降維 18
一、 主成分分析（Principal Component Analysis, PCA） 18
二、 奇異值分解（Singular Value Decomposition, SVD） 19
三、 堆疊自動編碼器（Stacked Auto-Encoders, SAE） 20
第三節 長短期記憶網絡（Long Short-Term Memory, LSTM） 22
第四節 雙向長短期記憶網絡（Bidirectional Long Short-Term Memory, Bi-LSTM） 23
第五節 計量指標 24
第五章 實證結果 26
第一節 資料分析 26
第二節 LSTM疫情前後相同模型相同產業之比較 30
第三節 Bi-LSTM疫情前後相同模型相同產業之比較 37
第六章 結論 39

Abedin, M. Z., Moon, M. H., Hassan, M. K., & Hajek, P. (2021). Deep learning-based exchange rate prediction during the COVID-19 pandemic. Annals of Operations Research, 1-52.
Allen, M. T., Madura, J., & Springer, T. M. (2000). REIT characteristics and the sensitivity of REIT returns. The Journal of Real Estate Finance and Economics, 21, 141-152.
Ariyo, A. A., Adewumi, A. O., & Ayo, C. K. (2014). Stock price prediction using the ARIMA model. 2014 UKSim-AMSS 16th international conference on computer modelling and simulation,
Assaf, A. (2015). Long memory and level shifts in REITs returns and volatility. International Review of Financial Analysis, 42, 172-182.
Bao, W., Yue, J., & Rao, Y. (2017). A deep learning framework for financial time series using stacked autoencoders and long-short term memory. PloS one, 12(7), e0180944.
Bengio, Y., Lamblin, P., Popovici, D., & Larochelle, H. (2006). Greedy layer-wise training of deep networks. Advances in neural information processing systems, 19.
Bentes, S. R. (2015). Forecasting volatility in gold returns under the GARCH, IGARCH and FIGARCH frameworks: New evidence. Physica A: Statistical Mechanics and its Applications, 438, 355-364.
Bloomberg. (2020). In Singapore, REITs Are Becoming More Important Than Ever. National Real Estate Investor: https://sg.finance.yahoo.com/news/in-singapore-rei-ts-are-becoming-more-important-than-ever-023113152.html
Bollen, J., Mao, H., & Zeng, X. (2011). Twitter mood predicts the stock market. Journal of computational science, 2(1), 1-8.
Cai, Y., & Xu, K. (2022). Net impact of COVID-19 on REIT returns. Journal of Risk and Financial Management, 15(8), 359.
Chen, Y., Lin, Z., Zhao, X., Wang, G., & Gu, Y. (2014). Deep learning-based classification of hyperspectral data. IEEE Journal of Selected topics in applied earth observations and remote sensing, 7(6), 2094-2107.
Choudhry, R., & Garg, K. (2008). A hybrid machine learning system for stock market forecasting. International Journal of Computer and Information Engineering, 2(3), 689-692.
Clayton, J., & MacKinnon, G. (2001). The time-varying nature of the link between REIT, real estate and financial asset returns. Journal of Real Estate Portfolio Management, 7(1), 43-54.
Ding, G., & Qin, L. (2020). Study on the prediction of stock price based on the associated network model of LSTM. International Journal of Machine Learning and Cybernetics, 11(6), 1307-1317.
Escher, M., Perneger, T. V., & Chevrolet, J.-C. (2004). National questionnaire survey on what influences doctors' decisions about admission to intensive care. BMJ, 329(7463), 425. https://doi.org/10.1136/bmj.329.7463.425
Ghosh, I., & Chaudhuri, T. D. (2021). FEB-stacking and FEB-DNN models for stock trend prediction: A performance analysis for pre and post Covid-19 periods. Decision Making: Applications in Management and Engineering, 4(1), 51-84.
Groth, S. S., & Muntermann, J. (2011). An intraday market risk management approach based on textual analysis. Decision Support Systems, 50(4), 680-691.
Hagenau, M., Liebmann, M., & Neumann, D. (2013). Automated news reading: Stock price prediction based on financial news using context-capturing features. Decision Support Systems, 55(3), 685-697.
Hao, Y., Chu, H. H., Ko, K. C., & Lin, L. (2016). Momentum strategies and investor sentiment in the REIT market. International Review of Finance, 16(1), 41-71.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
İpek, A. B. (2022). Stock price prediction using improved extreme learning machine methods during the Covid-19 pandemic and selection of appropriate prediction method. Kybernetes(ahead-of-print).
Ji, X., Wang, J., & Yan, Z. (2021). A stock price prediction method based on deep learning technology. International Journal of Crowd Science.
Jin, Z., Yang, Y., & Liu, Y. (2020). Stock closing price prediction based on sentiment analysis and LSTM. Neural Computing and Applications, 32(13), 9713-9729.
Jolliffe, I. T. (2003). Principal component analysis. Technometrics, 45(3), 276.
Kim, T., & Kim, H. Y. (2019). Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data. PloS one, 14(2), e0212320.
Kraus, M., & Feuerriegel, S. (2017). Decision support from financial disclosures with deep neural networks and transfer learning. Decision Support Systems, 104, 38-48.
Le, Q., & Mikolov, T. (2014). Distributed representations of sentences and documents Proceedings of the 31st International Conference on International Conference on Machine Learning - Volume 32, Beijing, China.
LEE, R. K. W., & Kam, T. S. (2014). Time-series data mining in transportation: A case study on singapore public train commuter travel patterns. International Journal of Engineering and Technology, 6(5), 431.
Mondal, P., Shit, L., & Goswami, S. (2014). Study of effectiveness of time series modeling (ARIMA) in forecasting stock prices. International Journal of Computer Science, Engineering and Applications, 4(2), 13.
Motegi, K., & Iitsuka, Y. (2022). Inter-regional Dependence of J-REIT Stock Prices: A Heteroscedasticity-Robust Time Series Approach. Available at SSRN 3902176.
Nassirtoussi, A. K., Aghabozorgi, S., Wah, T. Y., & Ngo, D. C. L. (2014). Text mining for market prediction: A systematic review. Expert Systems with Applications, 41(16), 7653-7670.
Nath S. (2023, April 24th). Medium. Why is BiLSTM better than LSTM ?: https://medium.com/@souro400.nath/why-is-bilstm-better-than-lstm-a7eb0090c1e4
Newell, G., Pham, A. K., & Ooi, J. (2015). The significance and performance of Singapore REITs in a mixed-asset portfolio. Journal of Property Investment & Finance, 33(1), 45-65.
Ngo, T. (2017). Exchange rate exposure of REITs. The Quarterly Review of Economics and Finance, 64, 249-258.
Olah C. (2015, August 27). Understanding LSTM Networks. colah's blog: https://colah.github.io/posts/2015-08-Understanding-LSTMs/
Ooi, J., Newell, G., & Sing, T.-F. (2006). The growth of REIT markets in Asia. Journal of Real Estate Literature, 14(2), 203-222.
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.
Raheem, I. D., Fasanya, I. O., & Yusuf, A. H. (2022). Economic policy uncertainty and the co-movement between REITs and exchange rate. Review of Economic Analysis, 14(1), 121-141.
Sajadi, S. M. A., Khodaee, P., Hajizadeh, E., Farhadi, S., Dastgoshade, S., & Du, B. (2022). Deep Learning-Based Methods for Forecasting Brent Crude Oil Return Considering COVID-19 Pandemic Effect. Energies, 15(21), 8124. https://www.mdpi.com/1996-1073/15/21/8124
Sen, J., & Chaudhuri, T. D. (2018). Stock price prediction using machine learning and deep learning frameworks. Proceedings of the 6th International Conference on Business Analytics and Intelligence, Bangalore, India,
Sen, J., Mehtab, S., & Nath, G. (2020). Stock price prediction using deep learning models. Lattice-The Machine Learning Journal, 1(3), 34-40.
Shen, J., & Shafiq, M. O. (2020). Short-term stock market price trend prediction using a comprehensive deep learning system. Journal of big Data, 7(1), 1-33.
Singapore REITs - REITSWEEK: List of REITs in Singapore. (2023, Agust 19). REITSWEEK: https://www.reitsweek.com/singapore-reits
Sunny, M. A. I., Maswood, M. M. S., & Alharbi, A. G. (2020). Deep learning-based stock price prediction using LSTM and bi-directional LSTM model. 2020 2nd novel intelligent and leading emerging sciences conference (NILES),
Vijh, M., Chandola, D., Tikkiwal, V. A., & Kumar, A. (2020). Stock closing price prediction using machine learning techniques. Procedia computer science, 167, 599-606.
Wang, H. (2020). Stock price prediction based on machine learning approaches. Proceedings of the 3rd International Conference on Data Science and Information Technology,
Weng, B., Lu, L., Wang, X., Megahed, F. M., & Martinez, W. (2018). Predicting short-term stock prices using ensemble methods and online data sources. Expert Systems with Applications, 112, 258-273.
Xie, X., Lu, Q., Herrera, M., Yu, Q., Parlikad, A. K., & Schooling, J. M. (2021). Does historical data still count? Exploring the applicability of smart building applications in the post-pandemic period. Sustainable Cities and Society, 69, 102804.
Xu, L., Magar, R., & Farimani, A. B. (2022). Forecasting COVID-19 new cases using deep learning methods. Computers in biology and medicine, 144, 105342. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8864960/pdf/main.pdf
Yu, Y., Duan, W., & Cao, Q. (2013). The impact of social and conventional media on firm equity value: A sentiment analysis approach. Decision Support Systems, 55(4), 919-926.
Yung, K., & Nafar, N. (2017). Investor attention and the expected returns of reits. International Review of Economics & Finance, 48, 423-439.