摘要：The purpose of this paper is to estimate bank’s failure by logistic regression and discrimant analysis. Both the logistic regression and discrimenant analysis showed that earning (E) was the most influential measure on banks’s failure followed by asset quality (A) liquidity (L) and capital adequacy (C). The estimated discrimenant function without cross validation obtained the following ratios 0.957, 0.872, 0.764, 1.000, 0.961 for fair, marginal, satisfactory, strong, and unsatisfactory respectively. While using cross validation it obtained 0.941, 0.872, 0.764, 1.000, and 0.961 respectively. Averages for the first and second method were 0.878 and 0.756 respectively. It is obvious that the estimated function without cross validation is the best for predicting fiscal situation of banks and the most efficient early warning system. A new bank is identified as being of a particular rating depends upon which discriminant function value is higher
