DEVELOPMENT OF A STABLE CORPORATE BANKRUPTCY CLASSIFICATION MODEL: EVIDENCE FROM TAIWAN
HONG LONG CHEN
This study constructs a corporate bankruptcy classification model with greater prediction accuracy that can be applied to a wide cross-section of industrial sectors. In Taiwan, development of a bankruptcy classification model for any one industry is difficult because of the small number of bankrupt companies per sector from it. Instead of using industry-relative ratios to stabilize the financial data, this study proposes an approach that combines financial ratio analysis and confirmatory factor analysis with logistic-regression analysis to estimate the probability of financial failure for public corporations. First, Mann-Whitney tests reveal a significant difference in the mean values of bankrupt and nonbankrupt companies for 41 financial ratios. Second, based on these financial ratios, a mathematical modeling procedure is used to develop bankruptcy classification model. Finally, validation of the bankruptcy model is by out-of-sample Type I accuracy, Type II accuracy, and overall correct classification rates. The research results suggest that the proposed modeling approach appears to be robust and relatively insensitive to differential industry effects and time variations.
Bankruptcy; Financial failure; Financial management; Logit models
HONG LONG CHEN (2018). Development of a stable corporate bankruptcy classification model: Evidence from Taiwan. International Journal of Economic Sciences, Vol. VII(1), pp. 16-38. , DOI: 10.52950/ES.2018.7.1.002
Copyright © 2018, Hong Long Chen et al, email@example.com