中文摘要：

传统的期货VaR风险度量模型缺乏对流动性风险的考虑。本文以变现成本为依据构建期货流动性指标，将流动性风险纳入VaR模型的研究，并结合中国期货市场实际特点。以上海金属铜、郑州硬麦和大连大豆为样本运用极值理论进行了实证研究，结果表明，农产品期货硬麦、大豆的流动性风险大于工业品金属铜期货的流动性风险；并且在衡量我国期货保证金设定时发现，对于我国期货市场，流动性风险是一个不可忽略的风险组成部分。为此。通过建立流动性风险调整后的保证金优化设置模型，使所设置的保证金水平能够涵盖整个期货市场的价格风险与流动性风险。

英文摘要：

Liquidity risk is one part of risks that could not be neglected in the futures market in China, but the classical VaR model is lack of consideration of liquidity risk. This paper holds that liquidity risk should be added into the research of VaR model, brings up a new method to measure risks after adjustment of liquidity risk, and constructs optimization model of setting up margins of futures market in China. Meanwhile, the paper applies POT model of Extreme Value Theory to make empirical study on liquidity risks of the futures market in China by using copper, wheat and soybean as samples. This method overcomes the shortage of modeling with whole sample. The empirical results show that the difference of liquidity risks of different futures is significant, the improved margin levels can cover price and liquidity risks of the futures market.