Background introduction
Credit loan refers to a loan product provided by banks or other financial institutions to small, medium and micro enterprises and individual industrial and commercial households. The characteristic of this loan is that it does not require the provision of collateral or guarantees, and is mainly evaluated and approved based on the borrower's credit status.
The application process for credit loans for small, medium and micro enterprises and individual industrial and commercial households is relatively simple. Applicants only need to provide relevant personal and corporate certification materials, such as ID cards, business licenses, tax registration certificates, etc., to apply for loans. The loan amount generally ranges from tens of thousands to several million yuan, and the loan period is relatively short, usually within one year.
The interest rates on credit loans for small, medium and micro enterprises and individual industrial and commercial households are relatively high because banks need to evaluate and approve the borrower's credit status and also need to bear certain risks. However, compared with other loan products, the application process of this loan is simpler, which is suitable for small, medium and micro enterprises and individual industrial and commercial households to quickly obtain financial support and promote the development of enterprises.
policy encouragement
In order to achieve long-term macroeconomic development and absorb employment, the Chinese government has expanded credit support for private enterprises and small and micro businesses. This is the long-term guideline and policy of our government. It is also a future trend for banks to increase loans to small and micro enterprises and individual industrial and commercial households.
Case
Teacher Toby provided credit loan data sets for small, medium and micro enterprises and individual industrial and commercial households for scorecard modeling. Data set objects include small and micro enterprises, individual industrial and commercial households, farmers, transportation loans, engineering loans, real estate mortgage loans, etc. The data sample size is tens of thousands and has good statistical significance.
Overdue rate
The overdue situation of small, medium and micro enterprises refers to the situation where the enterprise fails to repay the loan principal and interest on time within the loan repayment period. Overdue payments may have a negative impact on a company's credit history and operating conditions, while also increasing the company's financial pressure.
Overdue repayment may cause banks or financial institutions to take some measures against the enterprise, such as charging penalty interest, limiting the enterprise's credit limit, freezing the enterprise's accounts, etc. Long-term overdue repayment may even lead to the bank filing a lawsuit in court and taking legal measures to recover the arrears. For small, medium and micro enterprises, late repayment may affect the company's credit record, make future financing more difficult, and may even cause the company's capital chain to break and operational difficulties. Therefore, small, medium and micro enterprises should plan and manage funds reasonably to ensure repayment of loans on time and avoid overdue situations. At the same time, if the company does encounter operational difficulties, it should communicate with banks or financial institutions in a timely manner to seek reasonable solutions.
The overdue rate of this data set is about 30%, which is relatively high. The credit of small, medium and micro enterprises and individual industrial and commercial households is considered subprime loans, and banks and financial institutions need to bear greater risks. However, small, medium and micro enterprises and individual industrial and commercial households are the main force in the economy and employment, and banks and the government have to support them.
How to select high-quality loans to small, medium and micro enterprises and individual industrial and commercial households is a difficult problem for bank bosses.
Teacher Toby helps bank bosses solve problems by establishing credit score cards, automatically screening high-quality customers, and establishing credit scores for small, medium and micro enterprises and individual industrial and commercial households.
Data at a glance
As shown in the figure below, this data set has tens of thousands of entries and dozens of variables. The sample size is sufficient to achieve good statistical significance.
Portraits of loan users for small, medium and micro enterprises and individual businesses enumerate their ages. The age distribution is mainly between 40-50 years old. After all, he is a business person who has accumulated enough connections, experience and funds. It is more suitable for the business to start a business at the age of 40-50.
Teacher Toby processed the original data layer by layer, then divided it into bins, and then converted it into woe data. Binning smoothes the original data and reduces the variance of the variables. woe data is used for model training.
Model performance
The AUC of the scorecard model is close to 0.8. This is only a preliminary experimental result. Through multi-algorithm comparison, parameter adjustment, and variable screening, there is still room for improvement in model performance.
ks is greater than 0.4, and the model has good ability to distinguish good and bad customers.
credit score
In the United States, a FICO score, often called a credit score, is a three-digit number used to evaluate a person's likelihood of repaying credit when getting a credit card or loan from a lender. FICO scores are also used to help determine the interest rate on any credit extended to an individual. FICO scores range from 300 to 850 (worst to best). Every year, FICO scores (Fair, Isaac and Company) are widely used by various financial institutions and organizations. It can be said to be a very important criterion for judging a person's creditworthiness. Whether it is the success of the loan or the interest rate and discounts of the loan, they are closely related to your FICO credit score. In fact, 90% of financial institutions refer to FICO scores to make decisions, which illustrates the importance of FICO scores.
After our scorecard model for small, medium and micro enterprises and individual industrial and commercial households is established, we can also follow suit to establish FICO scores. The distribution of credit scores for small, medium and micro enterprises and individual industrial and commercial households is as follows, mainly concentrated around 600 points.
We can provide sufficient explanation for the variables, for example, the following is the calculation of the SHAP value of the loan period.
The longer the loan period, the greater the user's default risk; the shorter the loan period, the smaller the user's default risk.
This is consistent with the Monte Carlo algorithm simulation results of the previous article "Mathematical Traps Behind Long-Term Mortgage Loans", and it is also logically meaningful .
This is where we introduce to you the credit scorecard model and user portraits based on small, medium and micro enterprises and individual industrial and commercial households. If you are interested in this project, such as papers, patents, bank modeling, corporate modeling, corporate research needs, you can leave a message to contact our company. We provide the company's formal invoices and project contracts.
Welcome to learn more knowledge about risk control scorecard modeling"Python Credit Scorecard Modeling (with code)", we provide Professional scorecard models and other knowledge can realize automated credit scoring functions, create financial risk control credit approval models, and reduce risks.
The author is Toby, the article is from the public account (python risk control model),Credit scorecard model and user portrait based on small, medium and micro enterprises_individual industrial and commercial households