Abstract: Credit rating division is to divide customers into groups according to their credit score or expected losses, and to rank individuals by their credit risk. This study constructs a goal programming model from two aspects of default risk and risk loss, and gives a novel method of credit rating division for Chinese listed companies. This paper makes innovative attempts in two aspects. First, it puts forward new credit rating constraints to ensure that more defaulting customers are distributed in lower grades, thus avoiding the irrationality of the existing researches that distribute more defaulting customers in higher grades. Second, through the relationship between Expected Loss and Exposure at Default, Probability of Default and Loss Given Default in Basel Accord, this paper approximates the Expected Loss of listed companies and provides a solution for the lack of default loss data of listed companies. The results show that 53 indicators, such as equity ratio, return on equity, audit opinion type and the year-on-year growth rate of cash supply in circulation, can be used to effectively predict the default status of listed companies. The research shows that the corresponding loss rates of nine credit grades, such as AAA, AA, A, BBB, BB, B, CCC, CC and C, are 0.01%, 0.05%, 0.16%, 0.23%, 0.80%, 1.23%, 1.35%, 10.10% and 66.90%, respectively, and this is in line with the fundamental requirement that the higher credit rating, the lower loss rate.

Key words: credit rating division, loss rate, default, weight of evidence, listed company