数据与计算发展前沿 ›› 2023, Vol. 5 ›› Issue (6): 20-30.

CSTR: 32002.14.jfdc.CN10-1649/TP.2023.06.003

doi: 10.11871/jfdc.issn.2096-742X.2023.06.003

• • 上一篇    下一篇

基于RFM模型的上市公司违规行为画像研究

徐静1(),袁慧2,*()   

  1. 1.北京联合大学管理学院，北京 100101
    2.北京物资学院，北京 101149
  • 收稿日期:2023-04-28 出版日期:2023-12-20 发布日期:2023-12-25
  • 通讯作者: 袁慧（E-mail: 541735779@qq.com）
  • 作者简介:徐静，北京联合大学管理学院，副教授，北京大学博士后，美国注册管理会计师，在国内外重要期刊及会议上发表学术论文20余篇，完成专著3部，研究成果曾获丝路书香工程重点翻译项目资助，主要研究方向为监管科技和大数据审计。
    本文负责框架设计和论文撰写。
    XU Jing is an associate professor in the School of Management of Beijing Union University, a postdoctoral fellow at Peking University, and an American certified management accountant. She has published more than 20 academic papers in international and domestic journals and 3 monographs. Her research output has been funded by the National Key Translation Program of the Silk Road Project. Her current research interests are regulatory technology and big data auditing.
    In this paper, she is responsible for research design and thesis writing.
    E-mail: gltxj@buu.edu.cn|袁慧，北京物资学院，讲师，博士，主要研究方向为管理信息系统、数据挖掘等。
    本文负责论文数据收集和处理。
    YUAN Hui, Ph.D., is a lecturer at Beijing Wuzi University. Her main research interests are management information systems and data mining.
    In this paper, she is responsible for data collection and processing.
    E-mail: 541735779@qq.com
  • 基金资助:
    北京市社会科学基金规划项目“大数据审计模式下财务报表审计线索发现研究”(21GLB015)

Research on The Violation Portrait of Listed Companies Based on RFM Model

XU Jing1(),YUAN Hui2,*()   

  1. 1. School of Management, Beijing Union University, Beijing 100101, China
    2. Beijing Wuzi University, Beijing 101149, China
  • Received:2023-04-28 Online:2023-12-20 Published:2023-12-25

RichHTML

9

PDF

67

摘要：

【目的】 立足分类监管理念，通过刻画违规上市公司的多维特征，辅助监管部门进行风险苗头识别和违法违规线索发现。【方法】 以因违法违规受罚的我国制造业上市公司为研究对象，引入RFM模型进行违规风险指数评价，在此基础上，从盈利能力、偿债能力、营运能力、分红能力、资本结构、公司治理6个维度进行系统聚类，进而对违规上市公司进行画像。【结果】 根据RFM分值，将违规上市公司划分为低风险类、中风险类、次高风险类和高风险类，不同风险等级的违规上市公司体现出不同的特征，较弱的营运能力、偿债能力和公司治理能力以及高资本结构往往意味着高违规风险，这一结果符合违规行为发生的内在逻辑。【结论】 监管部门应依据上市公司的违规风险及其特征分类施策，从而进一步提升监管精准度。

关键词: RFM模型, 违规行为, 分类监管, 系统聚类, 企业画像

Abstract:

[Objective] Based on the concept of classified supervision, this paper depicts the multidimensional characteristics of illegal listed companies so as to assist the regulatory authorities in identifying signs of risk and discovering clues of illegal activities. [Methods] Taking China's listed manufacturing companies punished for violations as the research object, the RFM model is introduced to evaluate the violation risk index to reflect their violation severity and regulatory concern. On this basis, a hierarchical cluster analysis is performed in six dimensions: profitability, solvency, operation ability, dividend ability, capital structure, and corporate governance, and the violation portrait of listed companies is drawn. [Results] According to the RFM score, the illegal listed companies are divided into four categories: low risk, medium risk, sub high risk, and high risk. The listed companies with different violation risk levels have different characteristics. Weak operating capacity, solvency and corporate governance capacity, and high capital structure often mean high violation risk, which is consistent with the internal logic of violations. [Conclusions] The regulatory authorities should take differentiated measures according to the violation risks and characteristics of listed companies so as to further improve the regulatory accuracy.

Key words: RFM model, violation, classified supervision, hierarchical clustering, enterprise portrait