Optimal credit rating index combination method based on Fisher-Score default identification ability maximization

Abstract

The invention, which belongs to the field of the credit service technology, provides an optimal credit rating index combination method based on Fisher-Score default identification ability maximizationso that a problem that a credit risk identification failure is cause because the overall default identification ability of the existing credit rating system is not maximized can be solved. With Fisher-Score default identification ability maximization of a credit score as an optimal index combination standard, an index reflecting information repetition is rejected and then Fisher Score values of all formed index combinations are compared by using a traversing method, and the index group with the largest Fisher Score value of the credit score default identification ability is selected as an optimal index combination. With the method provided by the invention, the overall Fisher Score default identification ability maximization of the credit rating system is ensured and a decision-making basis is provided for investors like a bank or an individual to identify the credit risk effectively.

Description

technical field [0001] The present invention provides a method for selecting an optimal index combination of a credit rating system, and specifically relates to a method for selecting an optimal index combination of credit rating based on the FisherScore default identification ability to maximize the default identification ability (Fisher Score) of the credit rating system, which belongs to Credit service technology field. Background technique [0002] Credit is a loan activity conditional on the repayment of principal and interest. The purpose of credit rating is to evaluate the customer's credit level and its corresponding default rate through the value and status of rating indicators. [0003] Credit ratings and the default rates corresponding to different credit ratings are the basis of financial asset pricing. Mistakes in rating results can mislead individuals, companies, banks and other investors in their investment behavior, and even cause disorder in the entire fina...

AI Technical Summary

Problems solved by technology

[0009] The above-mentioned patented technology for quantitatively selecting indicators deletes individual indicators one by one, which cannot guarantee that the remaining indicator combinations have the greatest ability to identify defaults

Because the default identification ability of a single indicator is poor, the default identification ability in combination is not necessarily bad; similarly, the default identification ability of a single indicator is strong, but the default identification ability in combination is not necessarily strong

Images