基于Copula-GARCH方法的LPM套期保值研究  

Hedging with LPM based on Copula-GARCH method

作　　者：梁建峰[1] 陈健平[1] 刘京军[1] 

机构地区：[1]中山大学岭南学院,中山大学经济研究所,广东广州510275

出　　处：《系统工程学报》2011年第5期636-641,共6页Journal of Systems Engineering

基　　金：国家社会科学基金资助项目(08CJY064);国家自然科学基金资助项目(70871124);国家杰出青年基金资助项目(70825002);中央高校基本科研业务费专项资金资助项目(09WKPY35);全国优秀博士学位论文作者专项基金资助项目(200504)

摘　　要：风险的下偏距(lower partial moment,LPM)是一种较好的风险测度,它弥补了方差度量中的双边风险的不足,并且放松了对二次效用函数的限制要求.因此,LPM测度被广泛应用于金融风险管理研究.套期保值是金融风险控制的重要方法之一,但由于资产收益的联合分布的不确定性,给应用LPM进行套期保值带来了困难.对此采用Copula-GARCH方法对即期与衍生品市场的资产收益序列进行拟合,并将LPM模型应用于人民币外汇市场套期保值实证研究,得到最优套期保值比率.进一步通过绩效比较发现,无论在NDF市场还是远期市场,LPM模型的套期保值绩效都要优于最小方差的套期保值绩效.The lower partial moment （LPM） model has been widely applied in the research of hedging invest- ment. LPM focuses on the lower partial risk, so it is better than the variance measure. However the uncertainty of the joint distribution of the assets returns brings difficulties in the research and applications ofLPM. This article introduces Copula-GARCH method to fit the joint distribution, and applies the LPM model to the hedg- ing in RMB foreign exchange markets. Furthermore, the comparison between LPM and minimum-variance methods is carried out. It turns out that hedging based on the LPM model provides a better performance than that based on the minimum variance model on both the forward market and the NDF market.

关 键 词：LPM模型 COPULA函数 套期保值 绩效指标 

分 类 号：F830.59[经济管理—金融学] F830.92