ABSTRACT
This article aims to improve the predictive ability of KMV model by distinguishing firm size. The evidence suggests that default point would vary with firm size. Using the method of particle swarm optimization, we obtain the optimal default point separately for large firms and small firms. Several statistical tests such as the model confidence set methodology show that our relatively tractable model is more likely to have the strongest predictive ability.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 KMV is the acronym for Kealhofer, McQuown, and Vasicek, who originally propose the KMV model.
2 E is equal to market capitalization calculated by multiplying a company’s shares outstanding by the market price of one share. For nontradable shares, we offer a price discount of 59.75% suggested by Zhang and Shi (Citation2016).
3 To save space, we do not report the similar results, which are available upon request.