The Recognition Research On Earnings Management Of Listed Companies

Fraud on accounting information of listed companies is a common phenomenon in global security market now days. Security market in our country was born in the course of turning the planning economy to the market economy, so the specialty of both economy turning and new emerging capital market decides the distance between our security market and a regulatory market ruled by market oriented economy. So the important thing we are facing is to recognize and value the earnings management of listed companies, upon the points of both theory and practice. This paper is going to research on how to value the earnings management.The content of chapter one analyzes the study of domestic and foreign literatures. The second chapter mainly introduces the reason and method of earnings management. The recognition indexes system is established in the third chapter. Cluster analysis of SPSS is used to choose the representative indexes. Then weight of single indexes is calculated by the method of principle component analysis in the fourth chapter. In the fifth chapter, this paper discusses the steps of the model of the fuzzy comprehensive valuation for earnings management. In the last chapter, case study is used to test the validity of the model. Conclusion is made in the seventh chapter.The main character and innovation: firstly, fuzzy comprehensive valuation ranks the extent of earnings management of listed companies, which breaks through the deficiency of model of mathematical statistics. Secondly, indexes of related transaction ratio and extraordinary gains and losses ratio are introduced and it makes the recognition indexes system more adaptive to the environment of China stock market. In the last, cluster analysis is used to choose the representative indexes, lowering the correlation and quantity of the different indexes.