Research On Optimization Of Customer Credit Risk Assessment Method For Z Financial Leasing Company

Compared with the United States and other western countries,China’s financial leasing business started late.Since the establishment of the first financial leasing company in China in 1981,after decades of development,the financial leasing industry in China has witnessed a significant growth in both the size of assets and the number of institutions,and the distribution of industries and the geographical scope involved have become increasingly broad.However,what does not match the rapid growth is the lack of credit risk management and control means of China’s financial leasing companies.The reason is that,on the one hand,the credit risk management system of China’s financial industry is not perfect,and on the other hand,the research on credit risk assessment of China’s financial leasing companies is insufficient.Therefore,based on the basic principle of credit risk of financial leasing business,combined with the problems of credit risk assessment methods of Z financial leasing company,this thesis focuses on the research of its credit risk assessment model,and optimizes the customer risk assessment model of Z financial leasing company on the basis of empirical research.The digital economy is the future development direction of the world and plays an important role in the development of the world economy.The application of big data and artificial intelligence in the digital economy has triggered a change in the production mode,which has had a huge impact on the production process and commodity exchange process.Based on rich and huge data resources,it has become the general trend of financial risk management to establish a risk assessment model through big data technology,and link the external data optimization model in real time to improve the interception success rate of defaulting enterprises.In the past,the risk management and decision-making of financial leasing companies were mainly based on subjective experience and supplemented by data support,which led to the low level of enterprise risk management.Now,financial leasing companies can use big data technology to quantify the risks of business operation,establish a comprehensive risk management system,and enhance their core competitiveness.In the past,the customer risk assessment model of Z Financial Leasing Company relied more on expert experience,which was not objective and accurate enough.Therefore,based on the customer credit risk assessment model,this thesisuses three machine learning methods in big data technology: Logistic regression,support vector machine,and decision tree to establish a more objective and accurate assessment model.In the empirical analysis,the four models have good performance,so these models have great potential to be used in practical work.Due to the problems of multi-dimensional sample data,inconsistent scale of characteristic variables,and unbalanced positive and negative samples,when processing the sample data,this thesisuses the data standardization method to deal with the problem of different scale of each characteristic variable of the data sample,this thesis uses SMOTE method to solve the problem of unbalanced positive and negative sample data,and uses random forest model to solve the problem of too many dimensions of sample data.These methods are derived from big data technology and are suitable for financial leasing companies to pre-process huge business data,which is beneficial to improving model performance.The research of this thesishas certain reference value for improving Z financial leasing company’s customer credit risk assessment ability and improving the level of credit risk management and control as well as for other financial leasing companies.