本研究主要以世界上二種具有指標性的原油商品期貨(西德州中級原油期貨與北海布蘭特原油期貨)以及全球交易量最大的黃金期貨(紐約商品期貨交易所的黃金期貨)為研究對象，利用門檻自我迴歸共整合模型，及動差門檻自我迴歸共整合模型，分析在油價平穩以及波動高漲的兩個不同期間中各商品間的共整合互動關係，除此之外，並進一步利用門檻誤差修正模型對每一種商品組合間之長短期非對稱互動關係進行觀察。實證結果發現，在各商品間皆存在長期均衡之關係，另外，當市場上存有訊息衝擊效應可能會使得價格偏離長期均衡時具有不對稱之調整過程。當油價處在相對平穩的階段時，黃金價格的起伏會隨著原油價格而變動，突顯出黃金面對原油價格上升造成之通貨膨脹危機的避險保值特性。而當油價處在波動高漲的區間時，黃金與原油間之關係則不若油價平穩時密切。另外也發現黃金期貨在面對處於同一交易所的西德州原油商品期貨時，其誤差修正調整速度會高過於面對北海布蘭特原油期貨時的情形，即商品交易所間的地緣關係及交易時間之差異會影響商品反映市場衝擊的速度。 This paper employs the threshold autoregressive model, momentum threshold autoregressive model and threshold error-correction model to investigate the relationships between each pair of two oil benchmark (i.e., West Texas Intermediate, Brent) and gold prices in the futures market. Additionally, we divide full sample period into two periods which represent various volatility of oil price to see how are the results in each period differ from each other. The results of cointegration test reveal that long-run equilibrium relationships exist between the prices of crude oils and gold. Moreover, the results from Granger-Causality tests based on corresponding threshold error-correction model clearly point out there is asymmetry in the adjustment process and causality relationship between these commodities. According to the results of causality test, we found gold were affected by two oil benchmark when the volatility of oil price is low and stable, but this relationship is not so strong when the volatility of oil price becomes higher and more unstable. The results also show that gold adjusts much more sluggishly when it faces WTI, probably because the samples of Brent and gold used in this paper are traded in exchanges located in different countries which have a different trading time.