基于可扩展商业报告语言XBRL的财务欺诈模糊判别

发布时间：2019-01-12 08:08

【摘要】：自国内的证券市场开始到现在,金融诈骗案时有发生,这严重影响了投资者对投资的信心和投资激情,并且也严重干扰着资本市场的正常运行。最近,股市持续低迷,证监会不断的提高上市公司融资的门槛,又加上很多的中小企业板刚刚创建。上市企业的财务欺诈将不会减少反而会随之渐渐增多。怎样找寻一个简单通用的方法来识别上市企业是否发生财务欺诈俨然已经成为大部分的投资者、证券监督部门、会计师事务所和其他相关部门的研究热点。本文通过对上市公司财务各项财务指标进行了规范性的研究,并最终总结了国内外研究的一些热点指标进行了分析,从中选出了几个具有明显特征的指标作为本文的研究对象,建立了一个适合我国财务报告的金融欺诈决策模型,以此来判断上市公司是否有欺诈行文存在。国外的上市公司由于存在的时间比较长,发展历程比较久远,因此其相应的退市机制也比较可靠,相关的研究也较深层次,其研究的方面主要是对财务欺诈预警模型的研究,并获得了一定的研究成就。相比较国内的相关研究比较晚,主要还偏重于对国外研究成就的分析和描述阶段,并且通过数学模型建立了一些财务欺诈识别模型,在实证性研究方面也比较少。本文主要分析了中西方研究成果和方法后,结合了我国上市公司的实际情况,大胆尝试创建了一种基于模糊决策的上市公司财务欺诈识别模型。研究的对象主要集中了一些经被证实为财务欺诈的上市公司以及一些非欺诈上市公司,通过实证研究证明了此方法的实用性和可靠性。笔者首先根据我国上市公司的财务欺诈动机进行了分析,通过分析得到哪些存在欺诈的上市公司往往是财务指标异常和财务状况不佳的一些公司。并从中找到了财务欺诈引起异常的财务指标。本文研究的对象主要是从XBRL中已经被揭露存在财务欺诈的公司和未进行欺诈的正常上市公司进行研究,将选出的财务指标对这些上市公司进行分析研究。最后通过对以上指标进行模糊权重加权,得到了最终的模糊决策模型来识别上市公司的财务欺诈。本文通过研究得到了非财务欺诈的上市公司和已经有财务欺诈的上市公司之间有明显的指标差异。因此建立这个识别模型可以有效判断上市公司是否有欺诈行为,但是光靠此模型,来判断公司是否存在欺诈行为并不能完全可靠,因为还存在一些其他的因素干扰,因此这个模型仅能作为判断该公司是否有欺诈行为存在,只能起到一个警示的作用,要想得到真正的结果需要结合多方面的信息,收集更多的资料以得到更加精确的结果。
[Abstract]:Since the beginning of the domestic securities market, financial fraud cases have occurred from time to time, which has seriously affected investors' confidence in investment and investment passion, and seriously interfered with the normal operation of the capital market. Recently, the stock market continued to downturn, the Securities Regulatory Commission continued to raise the threshold of financing listed companies, plus a lot of small and medium-sized enterprises have just been created. Financial fraud of listed enterprises will not be reduced, but will gradually increase. How to find a simple and general method to identify financial fraud in listed enterprises has become a hot research topic for most investors, securities supervision departments, accounting firms and other relevant departments. This paper makes a normative research on the financial indexes of listed companies, and finally summarizes some hot research indicators at home and abroad, from which several indicators with obvious characteristics are selected as the research object of this paper. A decision model of financial fraud is established to judge the existence of fraud in listed companies. Because the foreign listed companies exist for a long time and have a long history of development, the corresponding delisting mechanism is also relatively reliable, and the related research is also deeper. The main research aspect is the study of the financial fraud warning model. And has obtained the certain research achievement. Compared with the domestic related research, the main emphasis on the analysis and description of foreign research achievements, and through the mathematical model to establish some financial fraud identification model, but also less empirical research. This paper mainly analyzes the research results and methods of China and the West, combines the actual situation of listed companies in China, and attempts to establish a financial fraud identification model of listed companies based on fuzzy decision making. The research focuses on some listed companies which have been proved to be financial fraud and some non-fraudulent listed companies. The practicability and reliability of this method are proved by empirical research. Firstly, according to the motivation of financial fraud of listed companies in China, the author finds out which listed companies are usually some companies with abnormal financial indicators and poor financial situation. And found out the financial fraud caused by abnormal financial indicators. The object of this paper is to study the listed companies which have been exposed to the existence of financial fraud from the XBRL and the normal listed companies without fraud. The selected financial indicators will be selected to analyze and study these listed companies. Finally, the final fuzzy decision model is obtained to identify the financial fraud of listed companies by weighted fuzzy weights. In this paper, we find that there are significant differences between listed companies with non-financial fraud and listed companies with financial fraud. Therefore, the establishment of this identification model can effectively judge whether a listed company is engaged in fraud, but it is not entirely reliable to judge whether there is fraud in a company by this model alone, because there are some other factors interfering with it. Therefore, this model can only serve as a warning to judge the existence of fraud in the company. In order to get real results, it is necessary to combine various information and collect more information to get more accurate results.
【学位授予单位】：云南财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F232

【参考文献】