Font Size: a A A

Study On Reasons, Detection And Protection Of Accounting Information Distortion In Company

Posted on:2011-06-21Degree:MasterType:Thesis
Country:ChinaCandidate:W ShangFull Text:PDF
GTID:2189330332970490Subject:Agricultural Economics and Management
Abstract/Summary:PDF Full Text Request
Since China built the capital market, which was represented by stock market, in 1990s, the quality of accounting information was focused by various aspects. Only because the accounting information take great effect in corporate operation, investment and financing, defining of property right, it is valuable to study. This thesis introduce the definition of accounting information briefly and describe the character of accounting information, the paper also prescribe the subject of accounting information. After describing the appearance of accounting information distortion in China, and the performance and the deficiency of the quality control of accounting information, three mathematical models are built on game theory in principal-agent frame, which are the outside users'game, the shareholders game with the managers, and the accounting information promulgators'game. The first model describes the deficiency of mining the accounting data that has already issued from the corporate. The second model indicates that, when the managers game the shareholders, they have incentive and ability to maneuver the accounting information. The third model shows that, on the promulgator side, there is no incentive to promulgate the accounting information. All these three issues are responsible for the accounting information distortion. Then, after depicting the step of traditional financing analyze which inherently has its advancement and shortcoming, this article build two finance early-warming models, according on the data that was proclaimed by the companies. The first one is binary logistic model, the second one is artificial neural network model. The two models analyze the relationship of the accounting information distortion with the finance data. Compare these two mathematical models shows that the ANN is superior to the binary logistic in prediction. Then a feasible method is given to distinguish the fakes from the veridical ones. After these work, an incentive method is built up to protect the accounting information distortion on root. In the end of this article, nine measures or propositions in different aspect are brought forward; especially emphasize the function of incentive system.
Keywords/Search Tags:accounting information, game theory, incentive, binary logistic regression, artificial neural network
PDF Full Text Request
Related items