存货质押业务质物组合价格风险决策

The Price Risk Decision of Inventory Portfolio in Supply Chain Finance

为缓释现行供应链金融业务单一质物价格剧烈波动诱发的集中度风险,基于Markowitz风险分散理论,从金融时间序列一般规律出发,分析价格随机波动现货质物的收益率统计特征,模型化收益率序列尖峰厚尾、波动集聚性和自相关特性,建立刻画质物组合间非线性相关结构的二元Copula-GARCH族模型,研究不同秩相关系数下两组质物组合对数收益率间的条件相关性,通过样本外滚动预测方法进行动态组合VaR预测;构造模拟生成新质物组合的数据生成方法,拓展研究不同相关性对组合VaR的影响;提出长周期预测视角中考虑资金成本的动态质押率模型效率损失检验以及基于Kupiec模型精度检验全面回测模型。实证结果显示:捕捉质物收益率下尾部相关结构变化的Clayton-Copula能够合理预测两组真实质物组合风险和模拟质物组合风险,且能更好地发挥组合分散风险的能力。为风险限额管理下的商业银行提供了一种质物组合风险管理的新框架和新模式。

In order to mitigate concentration risk due to sharp fluctuations of price of single inventory in supply chain finance, this paper presents an application of Copula-GARCH model in the dynamic estimation of the inventory portfolios＇ VaR with rolling time window based on the principle of risk diversification of Markowitz＇ s. This Copula-GARCH could better depict the characteristics of the autocorre-lation, heteroskedasticity, leptokurtosis and fat-tails of the marginal distribution, but also the conditional dependence of the joint distri- bution. In addition, four groups of portfolios with different correlation are generated which provides the impact different dependence struc-ture on portfolios＇ VaR. At last, the back testing are performed which consist of Kupiec testing and the efficiency loss testing based on dynamic impawn rate considering fund cost from the perspective of long-term forecasting. The results show that Clayton-Copula could rea-sonably estimate the risk of portfolio and diversify risk markedly. In summary, Copula-GARCH models provide a new framework of man-aging the risk of portfolio in inventory financing practice for banks constrained by risk limitation.