期刊文献+

改进支持向量机在虚假财务报告识别中的应用 被引量:8

Improved Support Vector Machine Algorithm for Fraudulent Financial Statements Detection
原文传递
导出
摘要 针对虚假财务报告识别实证研究的不足,本文提出了一种新的支持向量机方法对公司财务报告的真伪进行判别。第一,针对两类训练样本存在的"重叠"问题,建立双隶属支持向量机模型,通过基于谱聚类方法的隶属度模型来确定样本点对于两类样本的隶属程度;考虑到人们对于两种判别错误的"厌恶程度差异",在模型训练时,对训练样本进行了"非对称"处理,来降低虚假财务报告未识别的错误率。第二,在模型的输入财务指标选择上,本文通过浮动顺序搜索算法得到了虚假财务报告识别的全局最优财务指标组合。第三,实证结果表明,改进模型的判别准确率和泛化能力显著优于普通的支持向量机和BP神经网络,而公司治理指标的加入会提升模型的判别能力。 Empirical researches for fraudulent financial statements still remain in the early stage. This paper develops an improved SVM algorithm for the detection of financial statements. First, to solve the "overlap" problem between two classes of training samples, we build a dual membership support vector machine model and use the membership model based on spectral clustering to calculate the membership value of each sample belonging to the two classes; considering the difference between people's aversion to the two types of classification error, we do the unbalanced classifier training to reduce the error rate of fraudulent financial statements detection. Second, in the selection of financial indicators, we use the floating sequential search model to obtain the global optimal portfolio of financial indicators. Third, the empirical results show that the discrimination accuracy and generalization ability of our improved model are significantly better than traditional support vector machine and BP neural network; furthermore, the addition of corporate governance indicators will increase the detection capabilities.
出处 《管理评论》 CSSCI 北大核心 2012年第5期144-153,共10页 Management Review
基金 国家自然科学基金重点项目(70821061) 第六届全国大学生创新创业训练计划项目
关键词 虚假财务报告 支持向量机 隶属度模型 非对称训练 指标组合选择 fraudulent financial statements, support vector machine, membership model, unbalanced classifier training, indicators selection
  • 相关文献

参考文献24

  • 1Persons O. Using Financial Statement Data to Identify Factors Associated with Fraudulent Financial Reporting[J]. Journal of Applied Business Research, 1995,11(3):38-46. 被引量:1
  • 2Beneish M. D. Detecting GAAP Violation: hnplications for Assessing Earnings Management among Firms with Extreme Financial Performance[J]. Journal of Accounting and Public Policy, 1997,16(3):271-309. 被引量:1
  • 3Beneish M. D. Incentives and Penalties Related to Earnings Overstatements that Violate GAAP[J]. The Accounting Review, 1999, 74(4):425-457. 被引量:1
  • 4Hansen J. V., McDonald J. B., Messier W. F., et al. A Generalized Qualitative-response Model and the Analysis of Management Fraud[J]. Management Seienee, 1996,42(7):1022-1032. 被引量:1
  • 5Lee T. A., Ingram R. W., Howard T. P. The Difference between Earnings and Operating Cash Flow as an Indicator of Financial Reporting Fraud[J]. Contemporary Accounting Research, 1999,16(4):749-787. 被引量:1
  • 6Bell T., Carcello J. A Decision Aid for Assessing the Likelihood of Fraudulent Financial Reporting[J]. Auditing: A Journal of Practice and Theory, 2000,9(1):169-178. 被引量:1
  • 7Green B. P., Choi J. H. Assessing the Risk of Management Fraud through Neural-network Technology[J]. Auditing: A Journal of Practice and Theory, 1997,16(1):14-28. 被引量:1
  • 8Fanning K., Cogger K. Neural Network Detection of Management Fraud Using Published Financial Data[J]. International Journal of Intelligent Systems in Accounting, Finance and Management, 1998,7(1):21-24. 被引量:1
  • 9Kirkos E., Spathis C., Manolopoulos Y. Data Mining Techniques for the Detection of Fraudulent Financial Statements[J]. Expert Systems with Applications, 2007,32(4):995-1003. 被引量:1
  • 10蒙肖莲,李金林,杨毓.基于概率神经网络的欺诈性财务报告的识别研究[J].数理统计与管理,2009,28(1):36-45. 被引量:6

二级参考文献80

共引文献852

同被引文献70

引证文献8

二级引证文献24

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部