供应链金融多期价格风险测度与贷款组合优化

发布时间：2017-12-30 18:14

本文关键词：供应链金融多期价格风险测度与贷款组合优化　出处：《合肥工业大学》2017年硕士论文　论文类型：学位论文

【摘要】：中小企业是我国国民经济的重要组成部分,在促进经济增长、解决就业问题等方面起着重要作用。然而,由于其存在规模小、资信低等问题,使得其难以从正规金融机构获得融通资金。为有效解决此问题,由金融机构联合物流企业共同开展的供应链金融服务应运而生。虽然供应链金融发展空间广阔,但其在快速发展的同时,其风险也在不断增大。供应链金融以自偿性贸易融资为前提,而质押物的价值会受到外界环境诸多因素的影响而发生不确定性的波动,严重危及供应链金融的稳定发展。如何测度和防范单一质押物的价格风险以及对混合质押物进行贷款组合优化,以规避贷款集中度风险,已成为金融机构和物流企业所关注的重点问题。为此,本文主要开展了以下两个方面的研究工作。(1)供应链金融单一质押物多期价格风险测度。结合神经网络分位数回归(QRNN)能充分揭示质押物收益的非对称与非线性特征、供应链金融业务的多期等特点以及GARCH模型能刻画其波动聚集性等优势,首次构建了基于QRNN+GARCH的供应链金融多期价格风险VaR测度方法;其次,基于似然比检验与平均相对误差给出了测度效果的评估方法;再次,为设定合理的多期质押率,给出了质押率有效性评估指标:风险不可控比率和效率损失率。最后,选取现货铝进行实证研究,结果表明:QRNN+GARCH方法明显优于GARCH模型,表现为准确性更高,更具效率和稳健性,且其所确定的多期动态质押率,能更好的降低效率损失。(2)供应链金融混合质押物多期贷款组合优化。考虑到供应链金融中呈现出的非对称与非线性等典型特征,以分位数回归拟合单个资产边缘分布、以Copula函数刻画资产间非线性关联关系,建立Copula-分位数回归方法。使用该方法,对供应链金融多期贷款收益率进行预测,进而通过优化传统Sharpe比率、广义Omega比率等,给出贷款组合优化方案。选取现货铝和铜作为研究对象,实证研究发现:在各贷款期限下,Copula-分位数回归方法均优于Copula-GARCH方法,具体表现在前者拥有更高的Sharpe比率和广义Omega比率。本文给出了供应链金融多期价格风险测度与贷款组合优化方法与实证结果,能够为保障供应链金融健康发展提供决策参考,有利于实现金融机构、物流企业、核心企业以及中小企业共同发展的多赢局面。
[Abstract]:As an important part of our national economy, SMEs play an important role in promoting economic growth and solving employment problems. However, because of its small scale, low credit and other problems. It makes it difficult to obtain financing from formal financial institutions. To effectively solve this problem, the supply chain financial services jointly developed by financial institutions and joint logistics enterprises came into being. Although supply chain finance has a broad development space. Supply chain finance is based on self-compensated trade financing, and the value of pledge will be affected by many external environment factors. It seriously endangers the stable development of supply chain finance. How to measure and prevent the price risk of single pledge and optimize the loan portfolio of mixed pledge in order to avoid the risk of loan concentration. Has become the focus of financial institutions and logistics enterprises. This paper mainly carried out the following two aspects of research work. 1) the multi-period price risk measurement of single pledge in supply chain finance. Combined with neural network quantile regression QRNN). It can fully reveal the asymmetric and nonlinear characteristics of pledge proceeds. The characteristics of supply chain financial business such as multi-period and GARCH model can describe its volatility clustering and other advantages. The method of multi-period price risk measurement of supply chain based on QRNN GARCH is constructed for the first time. Secondly, based on likelihood ratio test and average relative error, the evaluation method of measurement effect is given. Thirdly, in order to set a reasonable multi-period pledge rate, the evaluation index of the validity of pledge rate is given: risk uncontrollable ratio and efficiency loss rate. Finally, spot aluminum is selected for empirical research. The results show that the GARCH method is superior to the GARCH model in accuracy, efficiency and robustness, and the multi-period dynamic pledge rate determined by the method is higher than that of the GARCH model. Can better reduce the efficiency loss. 2) supply chain finance portfolio optimization of multi-period mortgage mixed pledge. Considering the asymmetric and nonlinear characteristics of supply chain finance. The edge distribution of a single asset is fitted by quantile regression, and the nonlinear correlation between assets is described by Copula function. The Copula- quartile regression method is established. The yield of multi-period loan in supply chain finance is forecasted, and then the traditional Sharpe ratio, generalized Omega ratio and so on are optimized. The optimization scheme of loan portfolio is given. The spot aluminum and copper are selected as the research objects. The empirical study finds that: in each loan period. Copula-quantile regression method was superior to Copula-GARCH method. The former has higher Sharpe ratio and generalized Omega ratio. This paper gives the method and empirical results of multi-period price risk measurement and loan portfolio optimization in supply chain finance. It can provide decision reference for ensuring the healthy development of supply chain finance, and help to realize the win-win situation of the common development of financial institutions, logistics enterprises, core enterprises and small and medium-sized enterprises.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F274;F832.4

【相似文献】