中国农产品期货市场有限套利的证据

Evidence of Limited Arbitrage in China's Agricultural Commodity Futures Market

通过构建期现差价、期期差价与持有成本变量之间的回归模型和VAR模型,利用2004年1月—2014年1月中国玉米、强麦、棉花、黄大豆二号期货与现货价格日序列资料进行的回归分析和Granger因果检验发现：50%以上的单个合约期现价差率与市场利率、合约到期时间的变动关系与套利决定论不符;14%~60%的单个合约期现价差率与合约到期时间不存在显著关联或反相关;玉米、棉花期货的连续合约期现价差率、期期价差率与市场利率变动反相关;市场利率既不是这四种农产品期货连续合约期现价差率变动的Granger原因,也不是期期价差率变动的Granger原因。这些证据表明中国农产品期货市场套利不充分,应更重视预期在农产品期货价格形成中的作用。

This paper constructed the regression model and the VAR model to exam the relationship between futures spot （ or futures futures） price difference and holding cost variables. Empirical test is made on China＇s corn, wheat, cotton, soybean futures from January 2004 to January 2014. The results show that more than 50% of individual contract the regression relationship between futures spot price difference rate and market interest rates or maturity time inconsistent with arbitrage pricing; 14% -60% of individual contract the futures spot price difference rate negative or none related to maturity time;Corn, cotton continuous contract futures spot price difference rate and futures futures price difference rate negative related to market interest rate. Market interest rate is neither the Granger cause of agricultural commodity futures spot price difference rate nor the Granger cause of futures futures price difference rate. These evidences show that China＇s agricultural commodity futures market arbitrage is not sufficient, and we should pay more attention to the function of anticipation in agricultural commodity futures pricing.