论文部分内容阅读
本文提出了一种新的设定股指期货动态保证金水平的APARCH-GPD模型。它结合了APARCH模型良好拟合收益序列集丛性和尖峰厚尾分布的能力,以及GPD分布充分拟合尾部残差的特点,可提供准确的VaR风险度量。实证结果表明,使用该模型估计沪深300股指期货的保证金水平,其风险覆盖效果明显优于APARCH-norm模型、APARCH-t模型和APARCH-GED模型。研究还发现,目前股指期货市场的保证金水平偏高,具有下调空间,且空头头寸面临的价格波动风险要小于多头,因此可以对不同头寸设定差异化的保证金水平。
This paper presents a new APARCH-GPD model that sets the dynamic margin level of stock index futures. It combines the ability of the APARCH model to well fit the cluster of the income series and the peak-thick tail distribution, and fully fits the tail residuals with the GPD distribution to provide an accurate VaR risk measure. The empirical results show that using this model to estimate the margin level of Shanghai-Shenzhen 300 index futures is more effective than the APARCH-norm, APARCH-t and APARCH-GED models. The study also found that the current stock index futures market margin level is high, with room for downward adjustment, and short positions facing the risk of price fluctuations is less than long, so different positions can be set different margin levels.