This study investigates whether CAMEL(S) ratios can be used to predict
bank failure. Based on the literature review, the study used 13 variables
representing CAMEL ratios, one representing sensitivity to market risk, and
one representing bank size. Most of the analysis was done using
multivariate logistic regression since it is more flexible and relatively free
of restrictions. To evaluate for consistency, multiple discriminant analysis
was also carried out. The results found that logistic regression in tandem
with multiple discriminant analysis could function as an early warning
system for identifying bank failure and as a complement to on-site
examination. The results suggest that the variables ECTA (adequacy ratio),
RORA (assets quality), ROA (management), OEOI (earnings), CBTD
(liquidity), and LGBS (bank size) are statistically significant in explaining
bank failure. Therefore, stakeholders should focus on these variables to
identify and solve banking problems.