A study of credit risk measurement models:Evidence from Chinese listed companies

Zhao, Jing (2022) A study of credit risk measurement models:Evidence from Chinese listed companies. [Dissertation (University of Nottingham only)]

[img] PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (1MB)


This paper attempts a comparative study of credit risk measurement models by comparing the traditional models commonly used by Chinese companies: the Z-score and the Logit model to select a credit risk measurement model with better predictive performance for predicting the bankruptcy of Chinese listed companies. Traditional credit risk models are now widely used in China and have been regarded as the classical credit risk measurement tool worldwide. In the Z-score model, Altman's five independent variables: working capital to total assets, retained earnings, EBITDA to total assets, equity to debt and operating income to total assets, are used to predict bankruptcy. In the Logit model, the four independent variables of gearing ratio, return on total assets, working capital to total assets ratio and retained earnings ratio were selected and regressed to measure the probability of default, and the return on total assets ratio was found to be one of the most important factors influencing bankruptcy. Overall, both models are valid for predicting bankruptcy in China, and the Logit model outperforms the Z-score model.

Item Type: Dissertation (University of Nottingham only)
Keywords: Credit risk, Z-score model, Logit model, Financial ratios
Depositing User: ZHAO, JING
Date Deposited: 21 Jun 2023 15:07
Last Modified: 21 Jun 2023 15:07
URI: https://eprints.nottingham.ac.uk/id/eprint/70144

Actions (Archive Staff Only)

Edit View Edit View