A Survival Analysis Model for Credit Risk Appraisal of Commercial Banks and an Empirical Study

SONG Xue-feng,YANG Chao-jun,XU Ren-zhong

DOI: https://doi.org/10.3969/j.issn.1009-9190.2006.11.007

2006-01-01

Abstract:That an enterprise in financial crisis is unable to repay the due loans constitutes the major source of risks to credit assets of commercial banks.Thus it is of paramount importance for commercial banks to predict a company's financial crisis with an appropriate credit risk appraisal model so as to keep such credit risk at bay.This paper studies listed companies in our country and establishes a credit risk appraisal model based on survival analysis using Du Pont approach.The model provides 86%,72% and 68% prediction correctness respectively for 1,2 and 3 year randomly selected samples.By comparing with Altman model and Ohlson model and through robustness test,it is found that this model which can use time sequence and demands no sample match demonstrates high medium-and-long term forecastability and robustness.These characteristics put commercial banks in good stead in terms of medium-and-long term credit risk management.