本研究探討機器學習演算法結合不同風險測度於台灣股票市場動態資產配置之應用成效。傳統均值－變異數架構在面對高維度金融資料時，常受限於多重共線性、參數估計誤差及線性假設等問題，難以有效捕捉金融市場中複雜之非線性關係與結構變化。此外傳統變異數風險衡量方式將上漲與下跌所產生之波動視為同等風險，未能充分反映投資人對損失風險之非對稱偏好。基於上述限制，藉由台灣股票市場為研究對象，蒐集總體經濟、市場評價、利率環境及市場交易等相關變數作為預測因子，分別建立普通最小平方法、彈性網路迴歸及隨機森林模型，針對市場超額報酬、總風險與下行風險進行樣本外預測，並將預測結果導入不同資產配置架構以計算最適投資權重。此外更進一步運用SHAP解釋方法分析各預測因子對模型決策之影響程度，以提升機器學習模型之可解釋性。研究結果顯示機器學習模型在台股市場相較於傳統線性模型，能更有效捕捉金融市場之非線性資訊與動態變化。然而預測能力之提升未必能直接轉化為投資績效改善，其關鍵在於風險衡量架構之選擇。本研究結果不僅延伸傳統投資組合理論於非線性金融環境下之應用，亦為機器學習技術與下行風險管理整合於資產配置決策提供研究依據與實務參考。

This study investigates the performance of integrating machine learning algorithms and different risk measures in dynamic asset allocation within the Taiwan stock market. The traditional mean-variance framework often suffers from multicollinearity, parameter estimation errors, and linear assumptions when dealing with high-dimensional financial data, failing to capture complex non-linear relationships and structural changes. To address these limitations, this study utilizes the Taiwan stock market as the research object and collects macroeconomics, market valuation, interest rate environments, and market trading variables as predictor features. We construct Ordinary Least Squares, Elastic Net, and Random Forest models for out-of-sample forecasting of market excess returns, total risk, and downside risk, subsequently deriving optimal investment weights under different asset allocation frameworks. Additionally, the SHAP method is applied to analyze factor impacts on model decisions, enhancing machine learning interpretability. The empirical results show that machine learning models capture non-linear information and dynamic market shifts more effectively than traditional linear models. However, improved forecasting accuracy does not automatically translate into superior investment performance; the critical determinant lies in the choice of the risk management framework. In conclusion, the findings not only extend traditional portfolio theory to non-linear financial environments but also provide empirical evidence and a practical reference for integrating machine learning and downside risk management into asset allocation decisions.

陳昱安（2020），資產配置基於集成學習的多因子模型－以台灣股市為例（國立政治大學, 碩士論文）。
Ang, A., & Bekaert, G. (2007). Stock return predictability: Is it there? The Review of Financial Studies, 20(3), 651–707.
Bawa, V. S., & Lindenberg, E. B. (1977). Capital market equilibrium in a mean-lower partial moment framework. Journal of Financial Economics, 5(2), 189–200.
Ben Salah, O., Ben Rejeb, A., & El-Aroui, M. A. (2018). Portfolio optimization under downside risk framework: Empirical evidence from European markets. Journal of Asset Management, 19(4), 253–269.
Bianchi, D., Büchner, M., & Tamoni, A. (2021). Bond risk premiums with machine learning. The Review of Financial Studies, 34(2), 1046–1089.
Black, F., & Litterman, R. (1992). Global portfolio optimization. Financial Analysts Journal, 48(5), 28–43.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
Campbell, J. Y., & Thompson, S. B. (2008). Predicting excess stock returns out of sample: Can anything beat the historical average? The Review of Financial Studies, 21(4), 1509–1531.
Campbell, J. Y., & Shiller, R. J. (1988). The dividend-price ratio and expectations of future dividends and discount factors. The Review of Financial Studies, 1(3), 195–228.
DeMiguel, V., Garlappi, L., & Uppal, R. (2009). Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? The Review of Financial Studies, 22(5), 1915–1953.
Donaldson, R. G., & Kamstra, M. (1996). Forecast combining with neural networks. Journal of Forecasting, 15(1), 49–61.
Estrada, J. (2007). Mean-semivariance behavior: Downside risk and capital asset pricing. International Review of Economics & Finance, 16, 169–185.
Estrada, J. (2008). Downside risk in practice: The mean-semivariance approach. Journal of Applied Finance, 18(1), 58–70.
Fama, E. F. (1965). The behavior of stock-market prices. The Journal of Business, 38(1), 34–105.
Goyal, A., & Welch, I. (2008). A comprehensive look at the empirical performance of equity premium prediction. The Review of Financial Studies, 21(4), 1455–1508.
Goyal, A., & Welch, I. (2021). Empirical equity premium prediction: A review. Journal of Financial and Quantitative Analysis, 56(7), 2385–2415.
Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. The Review of Financial Studies, 33(5), 2223–2273.
Harlow, W. V. (1991). Asset allocation in a downside-risk framework. Financial Analysts Journal, 47(5), 28–40.
Hoerl, A. E., & Kennard, R. W. (1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1), 55–67.
Hollstein, F., Prokopczuk, M., & Simen, C. W. (2025). The global premium for stock market predictability. Journal of Financial Economics, 152, 103–124.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–292.
Kelly, B., & Pruitt, S. (2013). Market expectations in the cross-section of asset prices. The Journal of Finance, 68(5), 1821–1856.
Khaidem, L., Saha, S., & Dey, S. R. (2016). Predicting the direction of stock market prices using random forest. arXiv preprint arXiv:1605.00003.
Krauss, C., Do, X. A., & Huck, N. (2017). Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500. European Journal of Operational Research, 259(2), 689–702.
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems 30 (NeurIPS 2017) (pp. 4768–4777). Curran Associates, Inc.
Mandelbrot, B. (1963). The variation of certain speculative prices. The Journal of Business, 36(4), 394–419.
Markowitz, H. (1952). Portfolio selection. The Journal of Finance, 7(1), 77–91.
Pinelis, M., & Ruppert, D. (2022). Machine learning portfolio allocation. The Journal of Finance and Data Science, 8, 35-54.
Rapach, D. E., & Zhou, G. (2013). Forecasting stock returns. In G. Elliott & A. Timmermann (Eds.), Handbook of Economic Forecasting (Vol. 2, pp. 328–383). Elsevier.
Rigamonti, D. (2020). Downside risk in portfolio optimization: A re-examination. International Review of Financial Analysis, 72, 101–115.
Rigamonti, D., & Lucivjanska, K. (2022). Semivariance and asset allocation: Empirics on international stock markets. Journal of Empirical Finance, 67, 45–62.
Ross, S. A. (1976). The arbitrage theory of capital asset pricing. Journal of Economic Theory, 13(3), 341–360.
Rossi, A. G. (2018). Predicting stock market returns with machine learning. Journal of Financial Economics, 128(1), 183-206.
Sharpe, W. F. (1964). Capital asset prices: A theory of market equilibrium under conditions of risk. The Journal of Finance, 19(3), 425–442.
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267–288.
Zou, H., & Hastie, T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 67(2), 301–320.