Using company financial indicator to develop prediction model for banking bankruptcy in West Java survey in banking sector in West Java

Abstract

The purpose of this study is to develop model applicable for predicting banking bankruptcy in West Java. The model is developed based on banking financial indicator(Bank Indonesia, 1999). The data are collected with survey method and analyzed with three steps: First, calculate 931 financial ratio from the financial statement of (general) banking in West Java. Second, make the classification between healthy bank and unhealthy bank. Finally, make the model in order to predict banking bankruptcy and do model validation. The result of this study is the development of two models applicable for predicting banking bankruptcy based on discriminant analysis tlnd logistic regression. Both models are significant with alpha 1%, 5%, and 10%. Logistic regression model is more accurate than discriminant analysis model. Although the accuracy of the prediction model is verified, the bank predicted to go bankrupt may still exist in the reality. This happens for instance because of the lapping practices by the bank. Accordingly, the bankruptcy act is difficult to be imposed to the bank.