First, internal factors. In fact, the credit status of users and customers mainly depends on repayment ability and personal credit. Any question will lead to insufficient comprehensive score. Among them, it is best for personal credit to have no overdue record. If there is, it cannot be overdue at present, and it cannot be a credit black account; In addition, don't have too many records of credit inquiry in the short term.

Second, external factors. With the central bank's efforts to rectify credit cards, the bank's risk control has been tightened, further raising the threshold of related businesses. If the credit conditions of customers who have been good to the bank before have not changed, the comprehensive score will decrease with the increase of the threshold, so it is normal to have insufficient score.

Scoring classification:

According to different applications, credit score can be divided into risk score, income score, responsiveness score, customer churn score, collection score, credit card issuance audit score, housing mortgage loan issuance audit score, credit line approval score and so on.

Comprehensive credit risk score-Pengyuan 800 points

At the end of April 2005, Pengyuanzheng Letter Co., Ltd. independently developed a personal comprehensive credit risk score-"Pengyuan 800", which officially provided credit rating inquiry service for credit reporting institutions and individuals.

"Pengyuan 800" makes a statistical analysis of personal credit information by establishing a mathematical model, predicts the possibility of future default risk, and comprehensively reflects personal credit status with a score.

The credit scoring system has six grades, ranging from 320 to 800. Every 80 grades, the personal credit status is quantified in detail, and each score corresponds to a default probability. The higher the score, the lower the risk of default.

Fraction composition:

The scoring model selects more than forty variables related to personal credit, which are divided into four categories: personal basic information, bank credit information, personal payment information and personal fund status. Among them, the weight of bank credit information is the largest, close to 50%, and the other three categories are roughly equal.

At present, only 25% of the total population have bank credit records in the database of the credit information system. Because bank credit information is the most important variable affecting personal credit status, for customers without bank credit records, the model selects other variables related to bank credit instead. In the future, with the gradual improvement of data, we will add more variables to the model to continuously improve the accuracy, precision and universality of the model.