本研究主要透過一些過往被證實會對股價報酬有所影響的因子，如價值因子、品質因子、低波動因子、規模因子、動能因子等，和一些常用的股價基本資料和技術指標做為特徵，並透過機器學習模型挑選出預期表現好的股票並以此來建構投資組合。然而以上特徵與財報相關資料為季資料與股價基本資料和技術指標則為日資料，因此本研究中參考Hsu (2021)針對時間序列的特徵工程建構方式將日資料轉換為季資料，並以此來大幅增加特徵數量。
研究結果表明透過這樣子的選股模型不論是應用於同產業間選擇股票建構投資組合，又或者是在不同產業間各自選擇表現好的股票建構投資組合選股績效都是優於元大台灣卓越50基金(0050)的。此外研究中也考慮將REITs納入投資組合中，並且比較納入REITs與原先純股票投組的差異性。結果顯示在投資組合中納入REITs可以降低投組的波動性，並且獲得比原先單純股票的投組更穩定的投資績效表現。

This study primarily utilizes various factors that have been previously confirmed to impact stock returns, such as value, quality, low volatility, size, momentum factors, etc., along with commonly used stock fundamental data and technical indicators as features. Through machine learning models, it selects stocks expected to perform well and constructs an investment portfolio based on this selection. However, while the aforementioned features and financial data are quarterly, stock fundamental data and technical indicators are daily. Therefore, following the approach proposed by Hsu (2021) for time-series feature engineering, daily data is transformed into quarterly data, significantly increasing the number of features.
The results indicate that employing such a stock selection model, whether applied to choosing stocks within the same industry to construct an investment portfolio or selecting well-performing stocks across different industries, outperforms the Yuanta Taiwan Top 50 ETF (0050). Furthermore, the study considers incorporating REITs into the investment portfolio and compares the differences between including REITs and the original pure stock portfolio. The findings suggest that including REITs in the portfolio can reduce portfolio volatility and achieve a more stable investment performance compared to the original pure stock portfolio.

1.Banz, R. W. (1981). The relationship between return and market value of common stocks. Journal of financial economics, 9(1), 3-18.
2.Beath, A. D. (2023). Asset Allocation and Fund Performance of 3.Defined Benefit Pension Funds in the United States Between 1998–2021. CEM Benchmarking.
4.Breiman, L. (2001). Random forests. Machine learning, 45, 5-32.
5.Chen, J., & Peiser, R. (1999). The risk and return characteristics of REITs 1993-1997. Real Estate Finance, 16, 61-68.
7.Fama, E. F., & French, K. R. (1992). The cross‐section of expected 8.stock returns. The Journal of Finance, 47(2), 427-465.
9.Ghosh, C., Miles, M., & Sirmans, C. F. (1996). Are REITS stocks? Real Estate Finance, 13, 46-53.
10.Glascock, J. L. (1991). Market conditions, risk, and real estate portfolio returns: Some empirical evidence. The Journal of Real Estate Finance and Economics, 4, 367-373.
11.Glascock, J. L., Lu, C., & So, R. W. (2002). REIT returns and inflation: perverse or reverse causality effects? The Journal of Real Estate Finance and Economics, 24, 301-317.
12.Glascock, J. L., Michayluk, D., & Neuhauser, K. (2004). The riskiness of REITs surrounding the October 1997 stock market decline. The Journal of Real Estate Finance and Economics, 28, 339-354.
13.Haugen, R. A., & Heins, A. J. (1972). On the evidence supporting the existence of risk premiums in the capital market. Available at SSRN 1783797.
14.Hsu, T.-Y. (2021). Machine learning applied to stock index performance enhancement. Journal of Banking and Financial Technology, 5(1), 21-33.
15.Jegadeesh, N., & Titman, S. (1993). Returns to buying winners and selling losers: Implications for stock market efficiency. The Journal of Finance, 48(1), 65-91.
16.Khaidem, L., Saha, S., & Dey, S. R. (2016). Predicting the direction of stock market prices using random forest. arXiv preprint arXiv:1605.00003.
17.Lee, S., & Stevenson, S. (2005). The case for REITs in the mixed-asset portfolio in the short and long run. Journal of Real Estate Portfolio Management, 11(1), 55-80.
18.Ling, D. C., & Naranjo, A. (1999). The integration of commercial real estate markets and stock markets. Real Estate Economics, 27(3), 483-515.
19.Liu, C. H., Hartzell, D. J., Greig, W., & Grissom, T. V. (1990). The integration of the real estate market and the stock market: some preliminary evidence. The Journal of Real Estate Finance and Economics, 3, 261-282.
20.Lohrmann, C., & Luukka, P. (2019). Classification of intraday S&P500 returns with a Random Forest. International Journal of Forecasting, 35(1), 390-407.
21.Lowy, J. M. (1999). Real estate investment trusts. Tax Management Real Estate, 15(5), 115-158.
Markowitz, H. (1952). Portfolio selection. Journal of Finance, 7 (1), 77-91.
22.Novy-Marx, R. (2013). The other side of value: The gross profitability premium. Journal of financial economics, 108(1), 1-28.
23.Podkaminer, E. (2017). Smart beta is the gateway drug to risk factor investing. Journal of Portfolio Management, 43(5), 130.
24.Pomorski, P., & Gorse, D. (2023). Improving Portfolio Performance Using a Novel Method for Predicting Financial Regimes. International Conference on Machine Learning, Optimization, and Data Science
25.Rasekhschaffe, K. C., & Jones, R. C. (2019). Machine learning for stock selection. Financial Analysts Journal, 75(3), 70-88.
26.Sowden, W., & Young, M. S. (1996). The impact of real estate investment manager mergers. Real Estate Review, 26, 7-13.
27.Wilson, H., Fabozzi, S., & Gordon, J. (2003). Why Real Estate? An Expanding Role for Institutional Investors. Journal of Portfolio Management.