在當前波動劇烈且高度聯動的金融市場中，投資人需要有效的避險策略以分散風險、穩定報酬。由於資產間的相關性與市場波動性具有時間變動特性，能夠捕捉這類動態關係的模型對提升避險績效至關重要。本研究旨在探討無母數核密度估計 (KDE) 與傳統動態條件相關 DCC-GARCH 模型於 ETF 避險策略中的表現差異，並進一步比較兩者在美國與台灣市場中的應用成效。本文以代表性之正向與反向 ETF 組合構建避險部位，分別估算避險效率 (HE)、條件風險價值 (CVaR) 與平均報酬等績效指標，並採用拔靴法 (Bootstrap) 進行統計顯著性檢定。實證結果顯示， 實證結果顯示，KDE 模型於兩市場皆展現顯著高於 DCC 的避險效率，且美國市場表現優於台灣市場，顯示市場成熟度與流動性對避險模型績效具有一定影響。儘管台灣市場中 KDE 模型之平均報酬表現相對較低，但其在風險控制上仍具一定優勢，凸顯無母數方法在高波動與結構不穩定市場中之應用潛力。研究結果不僅補足過往對 KDE 與 DCC 模型缺乏直接比較之文獻空白，亦提供投資人於不同市場環境下選擇避險模型之實務依據。

In today’s highly volatile and interconnected financial markets, investors require effective hedging strategies to diversify risk and stabilize returns. Given the time-varying nature of asset correlations and market volatility, models capable of capturing such dynamic relationships are essential for improving hedging performance. This study aims to examine the performance differences between the nonparametric Kernel Density Estimation (KDE) method and the traditional Dynamic Conditional Correlation GARCH (DCC-GARCH) model in ETF-based hedging strategies, and to further compare their effectiveness across the U.S. and Taiwanese markets. Representative combinations of long and inverse ETFs are used to construct hedged portfolios, which are evaluated using performance indicators such as Hedging Effectiveness (HE), Conditional Value-at-Risk (CVaR), and average return. Statistical significance is tested using the bootstrap method. The empirical results show that the KDE model consistently achieves significantly higher hedging effectiveness than the DCC model in both markets. Furthermore, the U.S. market outperforms the Taiwanese market overall, suggesting that market maturity and liquidity may influence hedging model performance. Although the KDE model yields relatively lower average returns in the Taiwanese market, it still demonstrates notable advantages in risk control, highlighting the potential of nonparametric methods in volatile and structurally unstable markets. This research not only fills the gap in the literature regarding the direct empirical comparison between KDE and DCC models but also provides practical insights for investors in selecting suitable hedging models under different market conditions.

林展源(2019). “反向型 ETF 與波動型 ETF 之避險績效–應用 Copula-GJR GARCH模型”. 國立政治大學國際經營與貿易學系碩士班學位論文, 1-46.
吳尚澄(2024). “股債市場波動之動態相關：基於VIX和MOVE指數的分量迴歸研究”. 國立政治大學國際經營與貿易學系碩士班學位論文, 1-41.
Andersen, T. G. and Bollerslev, T. (1998). Answering the skeptics: Yes, standard volatility models do provide accurate forecasts. International Economic Review, 39(4), 885–905.
Baillie, R. T. and Myers, R. J. (1991). Bivariate GARCH estimation of the optimal commodity futures hedge. Journal of Applied Econometrics, 6(2), 109–124.
Bartram, S. M., Taylor, S. J. and Wang, Y. H. (2007). The Euro and European financial market dependence. Journal of Banking & Finance, 31(5), 1461–1481.
Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(3), 307–327.
Bollerslev, T. (1990). Modelling the Coherence in Short-Run Nominal Exchange Rates: A Multivariate Generalized ARCH Model. The Review of Economics and Statistics, 72(3), 498–505.
Bollerslev, T., Chou, R. Y. and Kroner, K. F. (1992). ARCH modeling in finance: A review of the theory and empirical evidence. Journal of Econometrics, 52(1–2), 5–59.
Chang, C.-L., McAleer, M., and Tansuchat, R. (2011). Crude oil hedging strategies using dynamic multivariate GARCH. Energy Economics, 33(5), 912–923.
Chen, Y. C., Wang, Y. H. and Lee, M. C. (2021). Dynamic hedging effectiveness under heavy tails and asymmetry. Finance Research Letters, 38, 101509.
Cho, J. (E.), Lee, G. H., Lee, W., & Kim, B. (2023). Machine Learning Approach for Predicting U.S. ETFs’ Tracking Errors – Implications on U.S. Invested Fund. Available at SSRN 4726993.
Choi, J.E. and Shin, D.W. (2021). Nonparametric estimation of time varying correlation coefficient. J. Korean Stat. Soc., 50, 333–353.
Curcio, R. J., Anderson, R. I., & Guirguis, H. (2015). On the Use of Leveraged Inverse ETFs to Hedge Risk in Publicly Traded Mortgage Portfolios. The Journal of Index Investing, 6(3), 40-57.
Ederington, L. H. (1979). The hedging performance of the new futures markets. Journal of Finance, 34(1), 157–170.
Efron, B. (1979). Bootstrap methods: Another look at the jackknife. The Annals of Statistics, 7(1), 1–26.
Engle, R. F. (1982). Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50(4), 987–1007.
Engle, R. (2002). Dynamic Conditional Correlation: A Simple Class of Multivariate Generalized Autoregressive Conditional Heteroskedasticity Models. Journal of Business & Economic Statistics, 20, 339-350.
Engle, R. F., Lilien, D. M. and Robins, R. P. (1987). Estimating time varying risk premia in the term structure: The ARCH-M model. Econometrica, 55(2), 391–407.
Forbes, K. J. and Rigobon, R. (2002). No contagion, only interdependence: Measuring stock market comovements. Journal of Finance, 57(5), 2223–2261.
Glosten, L. R., Jagannathan, R. and Runkle, D. E. (1993). On the relation between the expected value and the volatility of the nominal excess return on stocks. Journal of Finance, 48(5), 1779–1801.
Hill, J. M. and Foster, G. O. (2009). Understanding risk and returns of leveraged and inverse funds: Examining performance over time. The Journal of Index Investing, 1(2), 42–57.
Johnson, L. L. (1960). The theory of hedging and speculation in commodity futures. Review of Economic Studies, 27(3), 139–151.
Jondeau, E., & Rockinger, M. (2006). The Copula-GARCH model of conditional dependencies: An international stock market application. Journal of International Money and Finance, 25(5), 827–853.
Kroner, K. F., and Sultan, J. (1993). Time-varying distributions and dynamic hedging with foreign currency futures. Journal of Financial and Quantitative Analysis, 28(4), 535–551.
Longin, F. and Solnik, B. (1995). Is the correlation in international equity returns constant: 1960–1990? Journal of International Money and Finance, 14(1), 3–26.
Lu, L., Wang, J. and Zhang, G. (2009). Long Term Performance of Leveraged ETFs. Available at SSRN 1344133.
Patton, A. J. (2006). Modelling Asymmetric Exchange Rate Dependence. International Economic Review, 47(2), 527-556.
Rockafellar, R. T. and Uryasev, S. (2000). Optimization of conditional value-at-risk. Journal of Risk, 2(3), 21–41.
Sklar, A. (1959). Fonctions de répartition à n dimensions et leurs marges. Publications de I’ Institut deStatistique de l’University de Paris(8), 229-231.
Welch, B. L. (1947). The generalization of “Student’s” problem when several different population variances are involved. Biometrika, 34(1/2), 28–35.