Detection of financial fraud using recursive partitioning techniques

Financial fraud and then the consequent bankruptcy of a business is one of the reasons for the current financial crisis. The development and application of computational tools in financial fraud detection and bankruptcy prediction has become quite a popular cross-disciplinary research area in recent times involving financial engineers, forensic accountants and statistical modellers. In this paper we provide a review and comparative analysis of prediction performance of a battery of financial engineering tools used in financial fraud detection. We have found Hybrid model performs better as compared to other models. The computational models we have posited and reviewed using the real data can be generally applied to detect such financial frauds.