Fraud risk determine on financial statement: An application in a bank with artificial neural network model

The credit risk is the head of risks that banks come up against, because one of the main functions of the banks is credit usage. In the usages of commercial credits based on fraudulent financial statements, credit risk can occur if banks can not ensure the repayment of credits completely or partially, and this is an important problem in terms of banks. So the accuracy and the reliability of the information provided from financial statements has a crucial importance in credit risk management.In this context the main purpose of this study is to provide determining the fraud risk in financial statements and in this way to prevent the credit risk that can be occur in banks.In this study, to predict and determine the fraud risk in financial statements, the artificial neural network (ANN) technology is used as a method.The scope of the research consists of the commercial and corporate customers of a Bank. The financial data of 289 firms, belonging to the year of 2007, (97 firms were in manipulator group and 192 firms were in control group) was analyzed, and an ANN model was developed.The ANN model that was developed in the context of research has highly successful results by estimating 90% of the fraud risk through financial statements. The findings of ANN model was compared with several other statistical methods such as probit models, logit models and discriminate analysis and the most successful results was achieved by the ANN model.