XBRL财务报告与会计账簿研究：标准改进与数据集成

发布时间：2019-06-24 12:34

【摘要】：XBRL（可扩展商业报告语言）已在世界越来越多的国家和组织得到应用，尽管取得一系列瞩目的成绩，但是当前XBRL的发展也遇到了瓶颈：一方面，XBRL报告分类标准仍需进一步改进；另一方面，企业内部数据还不能“直通”XBRL财务报告。标准的不完善以及信息之间的割裂抑制了企业采用XBRL技术的积极性与主动性，造成了XBRL应用和推广的缓慢。已有的XBRL研究缺少整体视角，大多集中于财务报告环节，导致研究过于片面。本文基于XBRL报告信息链的整体视角，涵盖了XBRL会计账簿与XBRL财务报告，这样可以避免对XBRL的研究过于集中于信息链的某一个环节，同时围绕着“如何改进信息链各环节报告的分类标准”、“如何对信息链各环节进行数据集成”这两个基本问题展开研究。这些问题的解决能促使XBRL技术真正融入企业应用，并将推进XBRL的应用更上一个层次。本文综合运用了经济学、管理学与信息科学理论，采用理论研究与实证分析的研究方法，并在研究的不同阶段采用了不同的研究方法，甚至是多种方法相结合。首先，对与本文相关的研究文献进行回顾与梳理，分析XBRL的研究与应用现状。其次，基于XBRL标准的报告信息链，对现有的建模方式进行比较，并对报告信息链不同环节的建模方式予以确定。同时，由于技术因素对于XBRL报告信息链采纳的关键作用，进一步具体分析了影响XBRL报告信息链采纳的技术因素。进而，引出了本文的两个核心问题——XBRL标准改进与XBRL数据集成。在标准改进方面，分为XBRL账簿分类标准改进与XBRL财务报告分类标准改进，并分别予以分析。而在数据集成方面，通过构建XBRL数据集成框架，并分别通过信息技术中的映射方法与领域本体方法，研究纵向XBRL会计账簿与财务报告的数据集成、以及横向XBRL财务报告的数据集成问题。从而系统的完成了报告信息链上的XBRL标准改进与数据集成研究。本文的主要结论包括：（1）为了适应中国的信息披露特点，需要构建XBRL账簿分类标准中国模块，在此结论基础上，本文尝试构建了中国XBRL账簿分类标准草稿国际XBRL账簿分类标准提供企业内部数据的一般标准，并期望各国能够直接采用该标准。然而其真实的中国适应性水平并未加以检验，是否与中国的账簿特征相一致也亟待验证。本文以会计核算领域的中国国家标准为基准，对国际XBRL账簿分类标准进行适应性评价。研究结果发现，尽管两个标准都采用了XML语法规范，同时XBRL账簿分类标准在数据转换方面呈现出较强的完备性水平，然而现有的XBRL账簿分类标准并不能完全适用于中国的实际应用。通过比较分析，最终选择在借鉴国际XBRL账簿分类标准的基础上，构建中国模块的方案，同时运用分类标准工程学的方法，依据具体的创建步骤，构建了分类标准草稿。（2）可以通过聚类分析的方法，挖掘行业信息披露的相似性，从而提高XBRL财务报告行业分类标准的扩展效率国家未来将颁布更多的XBRL财务报告行业扩展分类标准，然而由于行业分类较多，能否更有效率的进行行业分类标准扩展也成为一个问题。本文对行业的信息披露特征进行挖掘。首先以2011年上交所的815份XBRL报告实例为研究样本，通过将企业XBRL报告实例与相对应的上交所分类标准进行匹配。实证研究发现行业在分类标准使用上存在显著的行业差异，，但差异并非在所有行业之间都存在。紧接着对行业的信息披露进行系统聚类，最终将已有的21个行业归纳为7类，并且类别之间的差异具有统计意义。同属一类的行业可以同时进行行业分类标准扩展，并且可以考虑共用元素。这样能够极大地降低行业分类标准的扩展成本，提升行业分类标准的创建效率与标准质量。（3）可以以报告信息链的视角进行XBRL数据集成研究当前XBRL的应用并不完整，并出现了一定的XBRL信息孤岛现象，需要对XBRL数据进行集成研究。本文以报告信息链为视角，构建信息链上的XBRL数据集成框架，包括纵向的XBRL会计账簿与财务报告的数据集成、横向的XBRL财务报告之间的数据集成。首先通过借鉴国际XBRL账簿分类标准的汇总报告上下文数据模块，实现会计账簿与财务报告之间的贯通，从而为XBRL财务报告的自动生成提供可能性选择。其次通过构建基于统一的财务报告领域本体，为XBRL财务报告的数据集成提供本体基础，从而能够从根本上提升XBRL财务报告的数据集成效率。同时通过应用研究加以检验，结果显示XBRL数据集成方案具有一定的可行性。本文站在一个更全面的信息链视角对XBRL进行研究，主要创新点包括：（1）XBRL报告信息链研究的细化以XBRL报告信息链为视角的研究有助于避免对XBRL的研究过于集中在某一个过程或者环节，并且能够兼顾信息链上更多利益相关者的利益。以往XBRL技术采纳的研究更多的是从监管者角度分析，较少考虑XBRL技术的实施主体。通过对XBRL报告信息链的细化研究，进而分析信息链中的微观结构以及影响信息链采纳的技术因素。（2）对国际XBRL账簿分类标准进行了中国适应性评价目前关于国际XBRL账簿分类标准在具体国家的适应性研究比较少见，本文是首次尝试。基于中国会计核算数据接口标准，本文对国际XBRL账簿分类标准的适应性水平进行评价，并且提供了定量化指标。在评价的基础上，还对其应用提出了未来确实可行的改进方案并加以实现。（3）通过对行业信息披露特征的聚类分析来建议行业分类标准扩展已有文献研究表明XBRL的财务报告信息披露呈现出行业差异，但尚未有文献进一步分析这种差异是行业间普遍存在的或是仅由少数行业引起。通过研究发现信息披露的行业差异并不表现在所有行业之间，并且除了少数信息披露比较特别的行业之外，其他行业的信息披露呈现出一定的相似关系。进而，本研究尝试首次通过聚类分析的方法，将信息披露相近的行业归为一类，从而为行业分类标准的扩展提供一种更为有效的方案，丰富了分类标准构建理论。（4）提出了统一的财务报告领域本体方案已有的文献研究关注于单一XBRL分类标准的本体表示，这样的做法能够增强XBRL分类标准的语义，但并不能从根本上提升XBRL财务数据的集成效率。本文提出统一的财务报告领域本体，构建基于该领域本体的XBRL财务报告集成模型，并以国内的XBRL分类标准为实验，进行XBRL财务报告转换应用。
[Abstract]:XBRL (Extensible Business Reporting Language) has been applied in a growing number of countries and organizations in the world, despite a series of high-profile achievements, the current XBRL development also has a bottleneck: on the one hand, XBRL reporting classification standards still need to be further improved; on the other hand, The internal data of the enterprise cannot also "straight-through" the XBRL financial report. The non-perfection of the standard and the separation between the information inhibits the enthusiasm and initiative of the enterprise to adopt the XBRL technology, resulting in the slow application and extension of XBRL. The existing XBRL research lacks an overall perspective, most of which is concentrated in the financial reporting link, resulting in the study being too thin This paper, based on the overall view of the XBRL report information chain, covers the XBRL accounting account books and XBRL financial reports, so that the research on XBRL can be avoided to be too centralized in a certain link of the information chain, and the two basic problems of the "How to Improve the Classification Standard of each Link Report of the Information Chain" and the "How to integrate the data in each link of the information chain" are developed. research. The solution to these problems can drive XBRL technology to be truly integrated into the enterprise application and will advance the application of XBRL This paper uses the theory of economics, management and information science, and adopts the research method of theoretical research and empirical analysis, and adopts different research methods and even a variety of methods in different stages of the research. In this paper, a review and analysis of the research literature related to this paper are carried out, and the study of XBRL should be analyzed. Secondly, based on the report information chain of XBRL standard, the existing modeling method is compared, and the modeling method of the different links of the report information chain is given. At the same time, due to the key role of the technical factors in the XBRL report information chain, this paper further analyses the technology that influences the adoption of the XBRL report information chain. Intraoperative factors. Furthermore, the two core issues of the article _ XBRL standard improvement and the number of XBRL It is integrated. In terms of standard improvement, it is divided into XBRL account book classification standard improvement and XBRL financial reporting classification standard improvement, and is respectively In that aspect of data integration, through the construction of the XBRL data integration framework, the data integration of the longitudinal XBRL accounting account book and the financial report and the data set of the transverse XBRL financial report are studied through the construction of the XBRL data integration framework and by the mapping method and the domain ontology method in the information technology, respectively. Problem. The system completes the XBRL standard improvement and data set on the report information chain in the form of a study. The principal of this article It is concluded that: (1) In order to adapt to the characteristics of China's information disclosure, it is necessary to construct the XBRL account book classification standard Chinese module, On the basis of this conclusion, this paper tries to construct the XBRL in China The draft international XBRL account book classification standard for the accounting book classification standard provides general standards for enterprise internal data and expects countries to The standard can be applied directly. However, its true level of China's adaptability has not been tested, and whether it is in contact with China's account book In this paper, the international XBRL account books are classified as the basis of the Chinese national standard in the field of accounting. The research results show that, although the two standards adopt the XML syntax specification, while the XBRL account book classification standard presents a strong level of completeness in the data conversion, the existing XBRL account book classification standard is not fully applicable. It is used in the practical application of China. Based on the comparison and analysis, the scheme of the Chinese module is finally selected based on the reference of the international XBRL account book classification standard, and the method of the classification standard engineering is applied, and according to the specific creating step, The classification standard draft is constructed. (2) The similarity of the information disclosure of the industry can be excavated by the method of cluster analysis, so as to improve the XBRL financial report. The expansion of the industry classification standard will result in more XBRL financial reporting industry extension classification in the future The standard, however, can be more efficient in the industry due to the more classification of the industry The extension of the classification standard is also a problem. The industry's information disclosure features are mined. First of all,815 XBRL reports from the Shanghai Stock Exchange are used to study samples, and by comparing the enterprise XBRL reports to the relative The empirical study found that there were significant differences in the industry in the use of the classification standards, but the difference The differences do not exist between all the industries. Next, the system clustering is carried out on the information disclosure of the industry, and the existing 21 industries are finally classified into 7 categories, and The difference between categories is of statistical significance. The same type of industry can carry out the industry classification standard at the same time Expand, and can take into account common elements. This can greatly reduce the expansion cost of industry classification standards and improve the industry score class standard creation efficiency and standard quality. (3) can be used to report information The XBRL data integration is not complete with the view of the chain, and some XBRL information islands appear. This paper, based on the report information chain, constructs the XBRL data integration framework on the information chain, including the vertical XBRL accounting account book and the data integration of the financial report, and the horizontal Data integration between XBRL financial reports. First, through the use of the summary report context data module of the international XBRL account book classification standard, the through between the accounting account book and the financial report is realized, thus the XBR The automatic generation of L's financial reports offers the possibility of choice. Secondly, through the construction of a unified financial reporting domain ontology, it provides a body foundation for the data integration of XBRL financial reports, so that it can fundamentally Improve the data integration efficiency of XBRL financial reporting. At the same time, it is checked by application research, and the results show that X The BRL data integration scheme has some feasibility. This paper is in a more comprehensive information chain The research of XBRL is the main innovation point of the angle of view It includes: (1) the research of XBRL report information chain research can help to avoid the research of XBRL report information chain as a visual angle The study of XBRL is too focused on a process or Link, and can balance the interests of more stakeholders on the information chain. Previous XBRL technology has adopted more research On the basis of the analysis of the regulator's angle, the implementation subject of XBRL technology is seldom considered. Through the detailed study of the information chain of XBRL report, the analysis is further analyzed. The microstructure of the information chain and the technical factors that affect the adoption of the information chain. (2) China's adaptive evaluation of the international XBRL account book classification standard is currently on the international XBRL account book This paper is the first attempt based on the data interface standard of China's accounting data, which is based on the data interface standard of China's accounting. The adaptability level of the accounting book classification standard is evaluated, and quantitative indexes are provided. The basis of the evaluation At the same time, it also puts forward some feasible and feasible improvement plan for its application. (3) It is recommended that the industry classification standard be expanded by cluster analysis of the characteristics of the industry information disclosure The research of the existing literature indicates that the disclosure of the financial report of XBRL presents the industry difference, but has not yet There is a literature that further analyses that this difference is common or only caused by a small number of industries. The industry difference disclosed by the study found that information disclosure does not appear between all industries, and in addition to a few pieces of information In addition to the specific industry, the disclosure of information in other industries presents a certain similarity. The extension of the industry classification standard provides a more effective solution The paper has enriched the construction theory of the classification standard. (4) The paper points out that the existing literature research in the uniform financial reporting field is concerned with the body representation of the single XBRL classification standard, and the practice can be enhanced. The semantic of XBRL classification standard, but does not fundamentally improve the integration efficiency of XBRL financial data. In this paper, a unified financial report field body is presented, and the XBRL financial report set based on the domain ontology is constructed.
【学位授予单位】：上海交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：F232

【参考文献】