中国上市公司财务造假预测模型  

Predicting Accounting Fraud in China

作　　者：钱苹[1] 罗玫[1] 

机构地区：[1]清华大学经济管理学院,100084

出　　处：《会计研究》2015年第7期18-25,96,共8页Accounting Research

基　　金：国家自然科学基金项目"中国股市是否相信上市公司发布的业绩预告"(71173125)的资助

摘　　要：本文以1994-2011年沪深两市财务造假上市公司为观察样本,首次全面综合地检验了国内外研究中表征财务造假和盈余质量问题的特征指标和西方资本市场常用的财务造假预测模型中的变量,建立了适合中国市场的更为可靠并且实用性强的财务造假预测模型。通过筛选检验大量的变量,本文发现其他应收款、是否亏损、经营应计项、现金销售率、股票换手率波动率、股权集中度、机构投资者持股比率、是否再融资和股市周期是鉴别中国上市公司造假的关键变量,用这些指标建立的综合模型不仅具有简单易懂的实用性,而且在辨别国内造假公司的能力方面显著优于西方常用的Mscore和Fscore模型。本文的研究结果为资本市场投资者、分析师、审计师和相关监管机构识别中国上市公司财务欺诈的嫌疑提供了简单实用的可靠方法。We analyze a sample of firms that have been subject to enforcement actions by the China Securities Regulatory Commission for allegedly misstating their financial statements during the period of 1994 to 2011,to develop a model for predicting accounting fraud in China. We examine variables of eight categories that best describe the earnings quality and characteristics of Chinese misstating firms,and variables from models built by Beneish( 1999) and Dechow( 2011). Our model provides a more reliable and accurate prediction than models built by Beneish( 1999) and Dechow( 2011),and it utilizes variables such as other receivables,total accruals,the percentage of cash sales,the loss indicator,the refinancing indicator,volatility of monthly turnover,share concentration,institutional holdings,and the stock market cycle. This paper finds a practical approach for regulators,analysts,auditors,and all the market participants to detect fraud in China.

关 键 词：财务造假 预测模型 阀值 

分 类 号：F275[经济管理—企业管理] F832.51[经济管理—国民经济]