商业银行知识产权质押贷款风险预警研究

发布时间：2018-01-06 14:18

本文关键词：商业银行知识产权质押贷款风险预警研究　出处：《天津财经大学》2013年硕士论文　论文类型：学位论文

【摘要】：随着知识经济的迅猛发展,知识产权质押贷款为企业加速科技成果资本化、拓展融资渠道提供了新的途径。2008年以来,政府为促进该项业务的开展先后出台了各项办法与指导意见,扶持高新技术企业利用知识产权获得贷款,但作为该项业务的主角——商业银行却因无法控制知识产权自身带来的高风险而谨慎行事。这不仅不利于商业银行在知识经济时代中积极调整原有担保结构,抢占新市场,更不利于整个经济市场上知识产权资本的增值,对科技创新产生一定的负面效应。如何才能合理预测知识产权质押贷款的风险呢?文章针对商业银行面临的这一棘手问题展开关于知识产权质押贷款风险预警的研究。一般来说风险预警研究主要包括风险识别和风险评估两大部分,文章主要从四个方面进行详细阐述。第一,知识产权质押贷款风险理论主要由有限性理论、新经济增长理论、不完全契约理论、预期收入理论、信息不对称理论五大部分组成,这五大理论从不同角度为风险成因提供了理论支持。第二,从宏观和微观两个层面对知识产权质押贷款风险影响因素进行识别并建立了囊括准入性指标与判定性指标的风险预警指标体系。经过研究发现,知识产权自身的无形性与独特性是产生高风险的根源,同时知识产权的法律属性导致商业银行可能遭受不可逆转的损失。第三,通过对已有风险预警模型的对比发现,BP神经网络是一个具有较强的自适应性、自我学习能力和容错性的非线性动态模型,适合对知识产权质押贷款业务的风险进行评估。第四,文章采用主成分分析法对知识产权质押风险状况进行判定,借鉴著名的“3σ”法则确定警度标准,构建了基于LM算法的三层BP神经网络模型,并对上市商业银行的数据进行训练与测试,实验结果表明该模型的准确率达到95%,能够有效的为商业银行知识产权质押贷款业务的风险进行预警。
[Abstract]:With the rapid development of knowledge economy, intellectual property mortgage loan has provided a new way for enterprises to accelerate the capitalization of scientific and technological achievements and expand financing channels. Since 2008. In order to promote the development of the business, the government has issued various measures and guidance to support high-tech enterprises to use intellectual property rights to obtain loans. However, as the protagonist of this business, commercial banks act prudently because they cannot control the high risks brought about by intellectual property rights, which is not only unfavorable for commercial banks to actively adjust the original guarantee structure in the era of knowledge economy. Seizing the new market is not conducive to the appreciation of intellectual property capital in the whole economic market, and has a certain negative effect on scientific and technological innovation. How can we reasonably predict the risk of intellectual property mortgage loans? In this paper, commercial banks face the thorny problem of intellectual property mortgage loan risk warning research. Generally speaking, risk early warning research mainly includes two parts: risk identification and risk assessment. Firstly, the risk theory of intellectual property mortgage loan mainly consists of finiteness theory, new economic growth theory, incomplete contract theory and expected income theory. The information asymmetry theory consists of five parts. These five theories provide theoretical support for the causes of risk from different angles. Second. This paper identifies the risk factors of intellectual property mortgage loan from macro and micro levels and establishes a risk early warning index system which includes access index and judgment index. The intangibility and uniqueness of intellectual property itself is the source of high risk. Meanwhile, the legal attribute of intellectual property rights may lead to irreversible losses to commercial banks. By comparing the existing risk warning models, it is found that BP neural network is a nonlinear dynamic model with strong self-adaptability, self-learning ability and fault tolerance. It is suitable to evaluate the risk of intellectual property mortgage loan business. 4th, this paper uses principal component analysis method to judge the risk of intellectual property mortgage, and uses the famous "3 蟽" rule to determine the alarm standard. A three-layer BP neural network model based on LM algorithm is constructed, and the data of listed commercial banks are trained and tested. The experimental results show that the accuracy of the model reaches 95%. It can effectively warn the risk of intellectual property mortgage loan business of commercial banks.
【学位授予单位】：天津财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.4

【参考文献】