本文主要探討農產品（玉米、大豆、大豆油和小麥）與軟性商品（可可、咖啡、棉花和糖）期貨契約的避險策略。利用不同的動靜態策略來進行避險績效的比較，這些避險策略分別是傳統最小平方法(OLS)、移動樣本的最小平方法(rollover OLS)、固定條件相關係數模型(constant conditional correlation, CCC)與動態條件相關係 數模型(dynamic conditional correlation, DCC)。在極小化避險投資組合變異數的目標下，本研究發現考慮資產報酬數列條件異和資產間動態關係的DCC避險策略，可以有效改善農產品與軟性商品樣本外的避險績效。

This paper compares the hedging effectiveness of eight commodity futures (including corns, soybeans, soybean oil, wheat, cocoa, coffee, cotton, and sugar) based on the hedge ratios estimated from the conventional ordinary least squares (OLS) method, the rollover OLS method, the constant conditional correlation (CCC) model, and the dynamic conditional correlation (DCC) model. In the framework of minimizing hedging portfolio variances, we find that the hedging strategy of the DCC model, which explicitly considers heteroscedasticity and time-varying correlations between the spot and futures returns, outperforms the others in this study.