供应链融资业务中钢材质押贷款动态质押率设定的VaR方法

发布时间：2019-05-09 08:48

【摘要】：异于债券、股票等质押融资业务,存货质押业务动态质押的核心在于预测其长期价格风险。从分析存货质押市场收益率的统计特征出发,以场外现货交易为主的钢材((HRB335)日数据为例,建立能刻画钢材收益率序列异方差性和尖峰厚尾特性的VaR-GARCH(1,1)-GED模型。同时,提出置于多风险窗口下运用样本外预测未来质押期内钢材价格风险水平,给出厚尾分布下长期风险VaR的计算解析式,得出与银行风险承受能力相一致的质押率。进而,基于失效率法则建立长期风险的碰撞序列函数,回测多风险窗口下长期VaR值。实证分析显示,模型得到的质押率在控制好风险的同时降低了效率损失,为商业银行提供一种动态质押率的风险管理模式和框架。
[Abstract]:Unlike bond, stock and other pledge financing business, the core of inventory pledge business dynamic pledge is to predict its long-term price risk. Based on the analysis of the statistical characteristics of the return rate on the inventory pledge market, the VaR-GARCH (1,1)-GED model, which can characterize the heteroscedasticity and the peak-thick-tail characteristics of the yield sequence of steel, is established by taking the daily data of (HRB335), which is dominated by over-the-counter spot transactions, as an example. At the same time, under the multi-risk window, it is put forward to forecast the steel price risk level in the future by out-of-sample, and the analytical formula of long-term risk VaR under the thick tail distribution is given, and the pledge rate consistent with the bank's risk bearing capacity is obtained. Furthermore, the collision sequence function of long-term risk is established based on the failure rate rule, and the long-term VaR value under multi-risk window is measured back. The empirical analysis shows that the pledge rate obtained by the model not only controls the risk well but also reduces the loss of efficiency and provides a dynamic risk management model and framework for commercial banks.
【作者单位】：西南交通大学交通运输与物流学院;复旦大学金融研究院;同济大学经济管理学院;华夏银行成都分行;
【基金】：国家自然科学基金资助项目(71003082) 全国博士后基金资助项目(20080430602) 教育部博士点基金资助项目(200806131007) 四川省科技计划软科学资助项目(2010ZR0028) 中央高校基本科研业务费专项资金科技创新资助项目(SWJTU11CX081)
【分类号】：F830.5;F407.3;F224

【参考文献】