Capability of Models Support Vector Regression, Least Angle Regression and Adaptive Neural Fuzzy Inference System for Earnings Management

Many studies have been made on the factors affecting earnings management and the level of correlation among them. But the factors affecting prediction of earnings management is less considered. This study examines capability of models Support Vector Regression (SVR) , Least Angel Regression (LARS) and Adaptive Neural Fuzzy Network (ANFIS) for earnings management. The sample includes firms listed in Tehran Stock Exchange from 2006 to 2012. The findings show that Support Vector Regression (SVR) , Least Angel Regression (LARS) and Adaptive Neural Fuzzy Inference System (ANFIS) have the most capability in earnings management respectively.