动态保证金是国际期货保证金制度发展的趋势,既顾及交易所的风险态度和有效地控制市场风险、又考虑到投资者的资金成本与效率是合理设定保证金水平的出发点。以交易活跃度为原则,选取1999年1月4日-2008年4月1日上海期货交易所的铜、大连商品交易所的大豆、郑州商品交易所的硬麦期货报酬率连续序列为样本,基于极值POT模型,利用谱风险测度模型对期货价格极端波动下设定动态保证金水平做实证研究,然后做回溯测试,并与用VaR和ES方法设定的保证金水平及现行静态保证金水平进行比较。研究结果表明,用极值POT模型可以很好地描述期货价格极端波动下报酬率序列的尾部特征,选择交易所风险厌恶参数为0.005计算极值谱风险动态保证金水平能对期货价格极端波动下的实时风险进行有效的控制,这为合理设定保证金水平的研究提供了新思路和新方法。
Dynamic margin is a development trend of international margin system. Not only attitudes of exchanges towards risks and efficient, control of market, but also cost and efficiency of money should betaken into account when setting up of margin reasonably. Taking activeness of trade as principle, the paper selects return rate series of continued futures contracts of copper in SHFE, hard wheat in ZCE and soybean in DCE from January 4, 1999 to April 1, 2008 as samples, And then , based on POT model, we do empirical study on setting up dynamic margin by spectral risk measurement of extreme value under extreme volatility of futures price, make back-test, and compare the result with margin set up based on value at risk and expected shortfall, and margin carried out nowadays. The results show, POT model could describe well tail characteristics of futures return rate series under extreme volatility of futures price. Computing dynamic margin using risk reversion coefficient 0. 005 of exchange by spectral risk measurement of extreme value could control risk on time efficiently under extreme volatility of futures price. This provides a new idea and approach for research of setting up margin reasonably .